prompt
stringlengths 94
42.6k
| completion
stringlengths 6
120
| api
stringlengths 14
68
|
|---|---|---|
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Test int8 quantizated model on ImageNet.
Note:
* QAT simulate int8 with fp32, gpu only.
* Quantized use real int8, cpu only, a bit slow.
* Results may be slightly different between qat and quantized mode.
"""
import argparse
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument(
"-m",
"--mode",
default="qat",
type=str,
choices=["normal", "qat", "quantized"],
help="Quantization Mode\n"
"normal: no quantization, using float32\n"
"qat: quantization aware training, simulate int8\n"
"quantized: convert mode to int8 quantized, inference only",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
test_proc = dist.launcher(worker) if world_size > 1 else worker
test_proc(world_size, args)
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
model = models.__dict__[args.arch]()
if args.mode != "normal":
quantize_qat(model, qconfig=Q.ema_fakequant_qconfig)
if args.checkpoint:
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
if args.mode == "quantized":
quantize(model)
# Define valid graph
def valid_func(image, label):
model.eval()
logits = model(image)
loss = F.loss.cross_entropy(logits, label, label_smooth=0.1)
acc1, acc5 = F.topk_accuracy(logits, label, (1, 5))
if dist.is_distributed(): # all_reduce_mean
loss = dist.functional.all_reduce_sum(loss) / dist.get_world_size()
acc1 = dist.functional.all_reduce_sum(acc1) / dist.get_world_size()
acc5 = dist.functional.all_reduce_sum(acc5) / dist.get_world_size()
return loss, acc1, acc5
# Build valid datasets
logger.info("preparing dataset..")
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), | T.ToMode("CHW") | megengine.data.transform.ToMode |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
import argparse
import multiprocessing as mp
import os
import cv2
import megengine as mge
import megengine.data as data
import megengine.data.dataset as dataset
import megengine.data.transform as T
import megengine.jit as jit
import numpy as np
from tqdm import tqdm
from official.vision.segmentation.deeplabv3plus import DeepLabV3Plus
class Config:
DATA_WORKERS = 4
NUM_CLASSES = 21
IMG_SIZE = 512
IMG_MEAN = [103.530, 116.280, 123.675]
IMG_STD = [57.375, 57.120, 58.395]
VAL_BATCHES = 1
VAL_MULTISCALE = [1.0] # [0.5, 0.75, 1.0, 1.25, 1.5, 1.75]
VAL_FLIP = False
VAL_SLIP = False
VAL_SAVE = None
cfg = Config()
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
"-d", "--dataset_dir", type=str, default="/data/datasets/VOC2012",
)
parser.add_argument(
"-m", "--model_path", type=str, default=None, help="eval model file"
)
args = parser.parse_args()
test_loader, test_size = build_dataloader(args.dataset_dir)
print("number of test images: %d" % (test_size))
net = DeepLabV3Plus(class_num=cfg.NUM_CLASSES)
model_dict = | mge.load(args.model_path) | megengine.load |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
import argparse
import multiprocessing as mp
import os
import cv2
import megengine as mge
import megengine.data as data
import megengine.data.dataset as dataset
import megengine.data.transform as T
import megengine.jit as jit
import numpy as np
from tqdm import tqdm
from official.vision.segmentation.deeplabv3plus import DeepLabV3Plus
class Config:
DATA_WORKERS = 4
NUM_CLASSES = 21
IMG_SIZE = 512
IMG_MEAN = [103.530, 116.280, 123.675]
IMG_STD = [57.375, 57.120, 58.395]
VAL_BATCHES = 1
VAL_MULTISCALE = [1.0] # [0.5, 0.75, 1.0, 1.25, 1.5, 1.75]
VAL_FLIP = False
VAL_SLIP = False
VAL_SAVE = None
cfg = Config()
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
"-d", "--dataset_dir", type=str, default="/data/datasets/VOC2012",
)
parser.add_argument(
"-m", "--model_path", type=str, default=None, help="eval model file"
)
args = parser.parse_args()
test_loader, test_size = build_dataloader(args.dataset_dir)
print("number of test images: %d" % (test_size))
net = DeepLabV3Plus(class_num=cfg.NUM_CLASSES)
model_dict = mge.load(args.model_path)
net.load_state_dict(model_dict["state_dict"])
print("load model %s" % (args.model_path))
net.eval()
result_list = []
for sample_batched in tqdm(test_loader):
img = sample_batched[0].squeeze()
label = sample_batched[1].squeeze()
pred = evaluate(net, img)
result_list.append({"pred": pred, "gt": label})
if cfg.VAL_SAVE:
save_results(result_list, cfg.VAL_SAVE)
compute_metric(result_list)
def pad_image_to_shape(img, shape, border_mode, value):
margin = np.zeros(4, np.uint32)
pad_height = shape[0] - img.shape[0] if shape[0] - img.shape[0] > 0 else 0
pad_width = shape[1] - img.shape[1] if shape[1] - img.shape[1] > 0 else 0
margin[0] = pad_height // 2
margin[1] = pad_height // 2 + pad_height % 2
margin[2] = pad_width // 2
margin[3] = pad_width // 2 + pad_width % 2
img = cv2.copyMakeBorder(
img, margin[0], margin[1], margin[2], margin[3], border_mode, value=value
)
return img, margin
def eval_single(net, img, is_flip):
@ | jit.trace(symbolic=True, opt_level=2) | megengine.jit.trace |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
import argparse
import multiprocessing as mp
import os
import cv2
import megengine as mge
import megengine.data as data
import megengine.data.dataset as dataset
import megengine.data.transform as T
import megengine.jit as jit
import numpy as np
from tqdm import tqdm
from official.vision.segmentation.deeplabv3plus import DeepLabV3Plus
class Config:
DATA_WORKERS = 4
NUM_CLASSES = 21
IMG_SIZE = 512
IMG_MEAN = [103.530, 116.280, 123.675]
IMG_STD = [57.375, 57.120, 58.395]
VAL_BATCHES = 1
VAL_MULTISCALE = [1.0] # [0.5, 0.75, 1.0, 1.25, 1.5, 1.75]
VAL_FLIP = False
VAL_SLIP = False
VAL_SAVE = None
cfg = Config()
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
"-d", "--dataset_dir", type=str, default="/data/datasets/VOC2012",
)
parser.add_argument(
"-m", "--model_path", type=str, default=None, help="eval model file"
)
args = parser.parse_args()
test_loader, test_size = build_dataloader(args.dataset_dir)
print("number of test images: %d" % (test_size))
net = DeepLabV3Plus(class_num=cfg.NUM_CLASSES)
model_dict = mge.load(args.model_path)
net.load_state_dict(model_dict["state_dict"])
print("load model %s" % (args.model_path))
net.eval()
result_list = []
for sample_batched in tqdm(test_loader):
img = sample_batched[0].squeeze()
label = sample_batched[1].squeeze()
pred = evaluate(net, img)
result_list.append({"pred": pred, "gt": label})
if cfg.VAL_SAVE:
save_results(result_list, cfg.VAL_SAVE)
compute_metric(result_list)
def pad_image_to_shape(img, shape, border_mode, value):
margin = np.zeros(4, np.uint32)
pad_height = shape[0] - img.shape[0] if shape[0] - img.shape[0] > 0 else 0
pad_width = shape[1] - img.shape[1] if shape[1] - img.shape[1] > 0 else 0
margin[0] = pad_height // 2
margin[1] = pad_height // 2 + pad_height % 2
margin[2] = pad_width // 2
margin[3] = pad_width // 2 + pad_width % 2
img = cv2.copyMakeBorder(
img, margin[0], margin[1], margin[2], margin[3], border_mode, value=value
)
return img, margin
def eval_single(net, img, is_flip):
@jit.trace(symbolic=True, opt_level=2)
def pred_fun(data, net=None):
net.eval()
pred = net(data)
return pred
data = | mge.tensor() | megengine.tensor |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
import argparse
import multiprocessing as mp
import os
import cv2
import megengine as mge
import megengine.data as data
import megengine.data.dataset as dataset
import megengine.data.transform as T
import megengine.jit as jit
import numpy as np
from tqdm import tqdm
from official.vision.segmentation.deeplabv3plus import DeepLabV3Plus
class Config:
DATA_WORKERS = 4
NUM_CLASSES = 21
IMG_SIZE = 512
IMG_MEAN = [103.530, 116.280, 123.675]
IMG_STD = [57.375, 57.120, 58.395]
VAL_BATCHES = 1
VAL_MULTISCALE = [1.0] # [0.5, 0.75, 1.0, 1.25, 1.5, 1.75]
VAL_FLIP = False
VAL_SLIP = False
VAL_SAVE = None
cfg = Config()
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
"-d", "--dataset_dir", type=str, default="/data/datasets/VOC2012",
)
parser.add_argument(
"-m", "--model_path", type=str, default=None, help="eval model file"
)
args = parser.parse_args()
test_loader, test_size = build_dataloader(args.dataset_dir)
print("number of test images: %d" % (test_size))
net = DeepLabV3Plus(class_num=cfg.NUM_CLASSES)
model_dict = mge.load(args.model_path)
net.load_state_dict(model_dict["state_dict"])
print("load model %s" % (args.model_path))
net.eval()
result_list = []
for sample_batched in tqdm(test_loader):
img = sample_batched[0].squeeze()
label = sample_batched[1].squeeze()
pred = evaluate(net, img)
result_list.append({"pred": pred, "gt": label})
if cfg.VAL_SAVE:
save_results(result_list, cfg.VAL_SAVE)
compute_metric(result_list)
def pad_image_to_shape(img, shape, border_mode, value):
margin = np.zeros(4, np.uint32)
pad_height = shape[0] - img.shape[0] if shape[0] - img.shape[0] > 0 else 0
pad_width = shape[1] - img.shape[1] if shape[1] - img.shape[1] > 0 else 0
margin[0] = pad_height // 2
margin[1] = pad_height // 2 + pad_height % 2
margin[2] = pad_width // 2
margin[3] = pad_width // 2 + pad_width % 2
img = cv2.copyMakeBorder(
img, margin[0], margin[1], margin[2], margin[3], border_mode, value=value
)
return img, margin
def eval_single(net, img, is_flip):
@jit.trace(symbolic=True, opt_level=2)
def pred_fun(data, net=None):
net.eval()
pred = net(data)
return pred
data = mge.tensor()
data.set_value(img.transpose(2, 0, 1)[np.newaxis])
pred = pred_fun(data, net=net)
if is_flip:
img_flip = img[:, ::-1, :]
data.set_value(img_flip.transpose(2, 0, 1)[np.newaxis])
pred_flip = pred_fun(data, net=net)
pred = (pred + pred_flip[:, :, :, ::-1]) / 2.0
del pred_flip
pred = pred.numpy().squeeze().transpose(1, 2, 0)
del data
return pred
def evaluate(net, img):
ori_h, ori_w, _ = img.shape
pred_all = np.zeros((ori_h, ori_w, cfg.NUM_CLASSES))
for rate in cfg.VAL_MULTISCALE:
if cfg.VAL_SLIP:
img_scale = cv2.resize(
img, None, fx=rate, fy=rate, interpolation=cv2.INTER_LINEAR
)
val_size = (cfg.IMG_SIZE, cfg.IMG_SIZE)
else:
out_h, out_w = int(cfg.IMG_SIZE * rate), int(cfg.IMG_SIZE * rate)
img_scale = cv2.resize(img, (out_w, out_h), interpolation=cv2.INTER_LINEAR)
val_size = (out_h, out_w)
new_h, new_w, _ = img_scale.shape
if (new_h <= val_size[0]) and (new_h <= val_size[1]):
img_pad, margin = pad_image_to_shape(
img_scale, val_size, cv2.BORDER_CONSTANT, value=0
)
pred = eval_single(net, img_pad, cfg.VAL_FLIP)
pred = pred[
margin[0] : (pred.shape[0] - margin[1]),
margin[2] : (pred.shape[1] - margin[3]),
:,
]
else:
stride_rate = 2 / 3
stride = [int(np.ceil(i * stride_rate)) for i in val_size]
print(img_scale.shape, stride, val_size)
img_pad, margin = pad_image_to_shape(
img_scale, val_size, cv2.BORDER_CONSTANT, value=0
)
pad_h, pad_w = img_pad.shape[:2]
r_grid, c_grid = [
int(np.ceil((ps - cs) / stride)) + 1
for ps, cs, stride in zip(img_pad.shape, val_size, stride)
]
pred_scale = np.zeros((pad_h, pad_w, cfg.NUM_CLASSES))
count_scale = np.zeros((pad_h, pad_w, cfg.NUM_CLASSES))
for grid_yidx in range(r_grid):
for grid_xidx in range(c_grid):
s_x = grid_xidx * stride[1]
s_y = grid_yidx * stride[0]
e_x = min(s_x + val_size[1], pad_w)
e_y = min(s_y + val_size[0], pad_h)
s_x = e_x - val_size[1]
s_y = e_y - val_size[0]
img_sub = img_pad[s_y:e_y, s_x:e_x, :]
timg_pad, tmargin = pad_image_to_shape(
img_sub, val_size, cv2.BORDER_CONSTANT, value=0
)
print(tmargin, timg_pad.shape)
tpred = eval_single(net, timg_pad, cfg.VAL_FLIP)
tpred = tpred[
margin[0] : (tpred.shape[0] - margin[1]),
margin[2] : (tpred.shape[1] - margin[3]),
:,
]
count_scale[s_y:e_y, s_x:e_x, :] += 1
pred_scale[s_y:e_y, s_x:e_x, :] += tpred
pred_scale = pred_scale / count_scale
pred = pred_scale[
margin[0] : (pred_scale.shape[0] - margin[1]),
margin[2] : (pred_scale.shape[1] - margin[3]),
:,
]
pred = cv2.resize(pred, (ori_w, ori_h), interpolation=cv2.INTER_LINEAR)
pred_all = pred_all + pred
pred_all = pred_all / len(cfg.VAL_MULTISCALE)
result = np.argmax(pred_all, axis=2).astype(np.uint8)
return result
def save_results(result_list, save_dir):
if not os.path.exists(save_dir):
os.makedirs(save_dir)
for idx, sample in enumerate(result_list):
file_path = os.path.join(save_dir, "%d.png" % idx)
cv2.imwrite(file_path, sample["pred"])
file_path = os.path.join(save_dir, "%d.gt.png" % idx)
cv2.imwrite(file_path, sample["gt"])
def compute_metric(result_list):
"""
modified from https://github.com/YudeWang/deeplabv3plus-pytorch
"""
TP, P, T = [], [], []
for i in range(cfg.NUM_CLASSES):
TP.append(mp.Value("i", 0, lock=True))
P.append(mp.Value("i", 0, lock=True))
T.append(mp.Value("i", 0, lock=True))
def compare(start, step, TP, P, T):
for idx in tqdm(range(start, len(result_list), step)):
pred = result_list[idx]["pred"]
gt = result_list[idx]["gt"]
cal = gt < 255
mask = (pred == gt) * cal
for i in range(cfg.NUM_CLASSES):
P[i].acquire()
P[i].value += np.sum((pred == i) * cal)
P[i].release()
T[i].acquire()
T[i].value += np.sum((gt == i) * cal)
T[i].release()
TP[i].acquire()
TP[i].value += np.sum((gt == i) * mask)
TP[i].release()
p_list = []
for i in range(8):
p = mp.Process(target=compare, args=(i, 8, TP, P, T))
p.start()
p_list.append(p)
for p in p_list:
p.join()
class_names = dataset.PascalVOC.class_names
IoU = []
for i in range(cfg.NUM_CLASSES):
IoU.append(TP[i].value / (T[i].value + P[i].value - TP[i].value + 1e-10))
for i in range(cfg.NUM_CLASSES):
if i == 0:
print("%11s:%7.3f%%" % ("backbound", IoU[i] * 100), end="\t")
else:
if i % 2 != 1:
print("%11s:%7.3f%%" % (class_names[i - 1], IoU[i] * 100), end="\t")
else:
print("%11s:%7.3f%%" % (class_names[i - 1], IoU[i] * 100))
miou = np.mean(np.array(IoU))
print("\n======================================================")
print("%11s:%7.3f%%" % ("mIoU", miou * 100))
return miou
def build_dataloader(dataset_dir):
val_dataset = | dataset.PascalVOC(dataset_dir, "val", order=["image", "mask"]) | megengine.data.dataset.PascalVOC |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
import argparse
import multiprocessing as mp
import os
import cv2
import megengine as mge
import megengine.data as data
import megengine.data.dataset as dataset
import megengine.data.transform as T
import megengine.jit as jit
import numpy as np
from tqdm import tqdm
from official.vision.segmentation.deeplabv3plus import DeepLabV3Plus
class Config:
DATA_WORKERS = 4
NUM_CLASSES = 21
IMG_SIZE = 512
IMG_MEAN = [103.530, 116.280, 123.675]
IMG_STD = [57.375, 57.120, 58.395]
VAL_BATCHES = 1
VAL_MULTISCALE = [1.0] # [0.5, 0.75, 1.0, 1.25, 1.5, 1.75]
VAL_FLIP = False
VAL_SLIP = False
VAL_SAVE = None
cfg = Config()
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
"-d", "--dataset_dir", type=str, default="/data/datasets/VOC2012",
)
parser.add_argument(
"-m", "--model_path", type=str, default=None, help="eval model file"
)
args = parser.parse_args()
test_loader, test_size = build_dataloader(args.dataset_dir)
print("number of test images: %d" % (test_size))
net = DeepLabV3Plus(class_num=cfg.NUM_CLASSES)
model_dict = mge.load(args.model_path)
net.load_state_dict(model_dict["state_dict"])
print("load model %s" % (args.model_path))
net.eval()
result_list = []
for sample_batched in tqdm(test_loader):
img = sample_batched[0].squeeze()
label = sample_batched[1].squeeze()
pred = evaluate(net, img)
result_list.append({"pred": pred, "gt": label})
if cfg.VAL_SAVE:
save_results(result_list, cfg.VAL_SAVE)
compute_metric(result_list)
def pad_image_to_shape(img, shape, border_mode, value):
margin = np.zeros(4, np.uint32)
pad_height = shape[0] - img.shape[0] if shape[0] - img.shape[0] > 0 else 0
pad_width = shape[1] - img.shape[1] if shape[1] - img.shape[1] > 0 else 0
margin[0] = pad_height // 2
margin[1] = pad_height // 2 + pad_height % 2
margin[2] = pad_width // 2
margin[3] = pad_width // 2 + pad_width % 2
img = cv2.copyMakeBorder(
img, margin[0], margin[1], margin[2], margin[3], border_mode, value=value
)
return img, margin
def eval_single(net, img, is_flip):
@jit.trace(symbolic=True, opt_level=2)
def pred_fun(data, net=None):
net.eval()
pred = net(data)
return pred
data = mge.tensor()
data.set_value(img.transpose(2, 0, 1)[np.newaxis])
pred = pred_fun(data, net=net)
if is_flip:
img_flip = img[:, ::-1, :]
data.set_value(img_flip.transpose(2, 0, 1)[np.newaxis])
pred_flip = pred_fun(data, net=net)
pred = (pred + pred_flip[:, :, :, ::-1]) / 2.0
del pred_flip
pred = pred.numpy().squeeze().transpose(1, 2, 0)
del data
return pred
def evaluate(net, img):
ori_h, ori_w, _ = img.shape
pred_all = np.zeros((ori_h, ori_w, cfg.NUM_CLASSES))
for rate in cfg.VAL_MULTISCALE:
if cfg.VAL_SLIP:
img_scale = cv2.resize(
img, None, fx=rate, fy=rate, interpolation=cv2.INTER_LINEAR
)
val_size = (cfg.IMG_SIZE, cfg.IMG_SIZE)
else:
out_h, out_w = int(cfg.IMG_SIZE * rate), int(cfg.IMG_SIZE * rate)
img_scale = cv2.resize(img, (out_w, out_h), interpolation=cv2.INTER_LINEAR)
val_size = (out_h, out_w)
new_h, new_w, _ = img_scale.shape
if (new_h <= val_size[0]) and (new_h <= val_size[1]):
img_pad, margin = pad_image_to_shape(
img_scale, val_size, cv2.BORDER_CONSTANT, value=0
)
pred = eval_single(net, img_pad, cfg.VAL_FLIP)
pred = pred[
margin[0] : (pred.shape[0] - margin[1]),
margin[2] : (pred.shape[1] - margin[3]),
:,
]
else:
stride_rate = 2 / 3
stride = [int(np.ceil(i * stride_rate)) for i in val_size]
print(img_scale.shape, stride, val_size)
img_pad, margin = pad_image_to_shape(
img_scale, val_size, cv2.BORDER_CONSTANT, value=0
)
pad_h, pad_w = img_pad.shape[:2]
r_grid, c_grid = [
int(np.ceil((ps - cs) / stride)) + 1
for ps, cs, stride in zip(img_pad.shape, val_size, stride)
]
pred_scale = np.zeros((pad_h, pad_w, cfg.NUM_CLASSES))
count_scale = np.zeros((pad_h, pad_w, cfg.NUM_CLASSES))
for grid_yidx in range(r_grid):
for grid_xidx in range(c_grid):
s_x = grid_xidx * stride[1]
s_y = grid_yidx * stride[0]
e_x = min(s_x + val_size[1], pad_w)
e_y = min(s_y + val_size[0], pad_h)
s_x = e_x - val_size[1]
s_y = e_y - val_size[0]
img_sub = img_pad[s_y:e_y, s_x:e_x, :]
timg_pad, tmargin = pad_image_to_shape(
img_sub, val_size, cv2.BORDER_CONSTANT, value=0
)
print(tmargin, timg_pad.shape)
tpred = eval_single(net, timg_pad, cfg.VAL_FLIP)
tpred = tpred[
margin[0] : (tpred.shape[0] - margin[1]),
margin[2] : (tpred.shape[1] - margin[3]),
:,
]
count_scale[s_y:e_y, s_x:e_x, :] += 1
pred_scale[s_y:e_y, s_x:e_x, :] += tpred
pred_scale = pred_scale / count_scale
pred = pred_scale[
margin[0] : (pred_scale.shape[0] - margin[1]),
margin[2] : (pred_scale.shape[1] - margin[3]),
:,
]
pred = cv2.resize(pred, (ori_w, ori_h), interpolation=cv2.INTER_LINEAR)
pred_all = pred_all + pred
pred_all = pred_all / len(cfg.VAL_MULTISCALE)
result = np.argmax(pred_all, axis=2).astype(np.uint8)
return result
def save_results(result_list, save_dir):
if not os.path.exists(save_dir):
os.makedirs(save_dir)
for idx, sample in enumerate(result_list):
file_path = os.path.join(save_dir, "%d.png" % idx)
cv2.imwrite(file_path, sample["pred"])
file_path = os.path.join(save_dir, "%d.gt.png" % idx)
cv2.imwrite(file_path, sample["gt"])
def compute_metric(result_list):
"""
modified from https://github.com/YudeWang/deeplabv3plus-pytorch
"""
TP, P, T = [], [], []
for i in range(cfg.NUM_CLASSES):
TP.append(mp.Value("i", 0, lock=True))
P.append(mp.Value("i", 0, lock=True))
T.append(mp.Value("i", 0, lock=True))
def compare(start, step, TP, P, T):
for idx in tqdm(range(start, len(result_list), step)):
pred = result_list[idx]["pred"]
gt = result_list[idx]["gt"]
cal = gt < 255
mask = (pred == gt) * cal
for i in range(cfg.NUM_CLASSES):
P[i].acquire()
P[i].value += np.sum((pred == i) * cal)
P[i].release()
T[i].acquire()
T[i].value += np.sum((gt == i) * cal)
T[i].release()
TP[i].acquire()
TP[i].value += np.sum((gt == i) * mask)
TP[i].release()
p_list = []
for i in range(8):
p = mp.Process(target=compare, args=(i, 8, TP, P, T))
p.start()
p_list.append(p)
for p in p_list:
p.join()
class_names = dataset.PascalVOC.class_names
IoU = []
for i in range(cfg.NUM_CLASSES):
IoU.append(TP[i].value / (T[i].value + P[i].value - TP[i].value + 1e-10))
for i in range(cfg.NUM_CLASSES):
if i == 0:
print("%11s:%7.3f%%" % ("backbound", IoU[i] * 100), end="\t")
else:
if i % 2 != 1:
print("%11s:%7.3f%%" % (class_names[i - 1], IoU[i] * 100), end="\t")
else:
print("%11s:%7.3f%%" % (class_names[i - 1], IoU[i] * 100))
miou = np.mean(np.array(IoU))
print("\n======================================================")
print("%11s:%7.3f%%" % ("mIoU", miou * 100))
return miou
def build_dataloader(dataset_dir):
val_dataset = dataset.PascalVOC(dataset_dir, "val", order=["image", "mask"])
val_sampler = | data.SequentialSampler(val_dataset, cfg.VAL_BATCHES) | megengine.data.SequentialSampler |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = | dtype.qint8(scale) | megengine.core.tensor.dtype.qint8 |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = | F.round(x) | megengine.functional.round |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = | F.clip(x, qmin, qmax) | megengine.functional.clip |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = | Float.QuantStub() | megengine.module.QuantStub |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = | QAT.QuantStub.from_float_module(normal_net) | megengine.module.qat.QuantStub.from_float_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
| disable_observer(qat_from_float) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
| disable_fake_quant(qat_from_float) | megengine.quantization.quantize.disable_fake_quant |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = | QAT.QuantStub() | megengine.module.qat.QuantStub |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
| disable_observer(qat_net) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
| propagate_qconfig(qat_net, min_max_fakequant_qconfig) | megengine.quantization.quantize.propagate_qconfig |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = | Q.QuantStub.from_qat_module(qat_net) | megengine.module.quantized.QuantStub.from_qat_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = | Float.DequantStub() | megengine.module.DequantStub |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = | QAT.DequantStub.from_float_module(normal_net) | megengine.module.qat.DequantStub.from_float_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
| disable_fake_quant(qat_from_float) | megengine.quantization.quantize.disable_fake_quant |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
| disable_observer(qat_from_float) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = | QAT.DequantStub() | megengine.module.qat.DequantStub |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
| disable_observer(qat_net) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
| propagate_qconfig(qat_net, min_max_fakequant_qconfig) | megengine.quantization.quantize.propagate_qconfig |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = | Q.DequantStub.from_qat_module(qat_net) | megengine.module.quantized.DequantStub.from_qat_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = | Float.Elemwise(kind) | megengine.module.Elemwise |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = | QAT.Elemwise.from_float_module(normal_net) | megengine.module.qat.Elemwise.from_float_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
| disable_observer(qat_from_float) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
| disable_fake_quant(qat_from_float) | megengine.quantization.quantize.disable_fake_quant |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = | QAT.Elemwise(kind) | megengine.module.qat.Elemwise |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
| disable_observer(qat_net) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
| propagate_qconfig(qat_net, min_max_fakequant_qconfig) | megengine.quantization.quantize.propagate_qconfig |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = | Q.Elemwise.from_qat_module(qat_net) | megengine.module.quantized.Elemwise.from_qat_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = | Float.Linear(3, 3, bias=True) | megengine.module.Linear |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = | QAT.Linear(3, 3, bias=True) | megengine.module.qat.Linear |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
| disable_observer(qat_net) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
| propagate_qconfig(qat_net, min_max_fakequant_qconfig) | megengine.quantization.quantize.propagate_qconfig |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = | Parameter(weight) | megengine.Parameter |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = | Parameter(bias) | megengine.Parameter |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = | QAT.Linear.from_float_module(normal_net) | megengine.module.qat.Linear.from_float_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
| disable_fake_quant(qat_from_float) | megengine.quantization.quantize.disable_fake_quant |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
| disable_observer(qat_from_float) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = | Q.Linear.from_qat_module(qat_net) | megengine.module.quantized.Linear.from_qat_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
| disable_observer(qat_net) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
| propagate_qconfig(qat_net, min_max_fakequant_qconfig) | megengine.quantization.quantize.propagate_qconfig |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
| disable_observer(qat_from_float) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
| disable_fake_quant(qat_from_float) | megengine.quantization.quantize.disable_fake_quant |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = | Float.Concat() | megengine.module.Concat |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = Float.Concat()
normal_net.eval()
qat_net = | QAT.Concat() | megengine.module.qat.Concat |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = Float.Concat()
normal_net.eval()
qat_net = QAT.Concat()
qat_net.eval()
| disable_observer(qat_net) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = Float.Concat()
normal_net.eval()
qat_net = QAT.Concat()
qat_net.eval()
disable_observer(qat_net)
| propagate_qconfig(qat_net, min_max_fakequant_qconfig) | megengine.quantization.quantize.propagate_qconfig |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = Float.Concat()
normal_net.eval()
qat_net = QAT.Concat()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
inps = []
inps_int8 = []
for i in range(3):
inp_scale = gen_inp_scale()
inps.append(mge.tensor(np.random.normal(size=(3, 3)).astype("float32")))
inps[i] = fake_quant_act(inps[i], inp_scale)
inps[i].qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
inps_int8.append(quant(inps[i], inp_scale))
qat_from_float = | QAT.Concat.from_float_module(normal_net) | megengine.module.qat.Concat.from_float_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = Float.Concat()
normal_net.eval()
qat_net = QAT.Concat()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
inps = []
inps_int8 = []
for i in range(3):
inp_scale = gen_inp_scale()
inps.append(mge.tensor(np.random.normal(size=(3, 3)).astype("float32")))
inps[i] = fake_quant_act(inps[i], inp_scale)
inps[i].qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
inps_int8.append(quant(inps[i], inp_scale))
qat_from_float = QAT.Concat.from_float_module(normal_net)
qat_from_float.eval()
| disable_fake_quant(qat_from_float) | megengine.quantization.quantize.disable_fake_quant |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = Float.Concat()
normal_net.eval()
qat_net = QAT.Concat()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
inps = []
inps_int8 = []
for i in range(3):
inp_scale = gen_inp_scale()
inps.append(mge.tensor(np.random.normal(size=(3, 3)).astype("float32")))
inps[i] = fake_quant_act(inps[i], inp_scale)
inps[i].qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
inps_int8.append(quant(inps[i], inp_scale))
qat_from_float = QAT.Concat.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
| disable_observer(qat_from_float) | megengine.quantization.quantize.disable_observer |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = Float.Concat()
normal_net.eval()
qat_net = QAT.Concat()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
inps = []
inps_int8 = []
for i in range(3):
inp_scale = gen_inp_scale()
inps.append(mge.tensor(np.random.normal(size=(3, 3)).astype("float32")))
inps[i] = fake_quant_act(inps[i], inp_scale)
inps[i].qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
inps_int8.append(quant(inps[i], inp_scale))
qat_from_float = QAT.Concat.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = | Q.Concat.from_qat_module(qat_net) | megengine.module.quantized.Concat.from_qat_module |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = | Tensor(min_val[0]) | megengine.Tensor |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = | Tensor(max_val[0]) | megengine.Tensor |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = | Tensor(min_val[1]) | megengine.Tensor |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = | Tensor(max_val[1]) | megengine.Tensor |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update( | create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale) | megengine.quantization.create_qparams |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update( | create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale) | megengine.quantization.create_qparams |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = | Parameter(weight) | megengine.Parameter |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = | Parameter(bias) | megengine.Parameter |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = | Parameter(weight) | megengine.Parameter |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = | Parameter(bias) | megengine.Parameter |
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from functools import partial
import numpy as np
import pytest
import megengine as mge
import megengine.functional as F
import megengine.module as Float
import megengine.module.qat as QAT
import megengine.module.quantized as Q
from megengine import Parameter, Tensor
from megengine.core.tensor import dtype
from megengine.quantization import (
FakeQuantize,
MinMaxObserver,
QConfig,
QuantMode,
create_qparams,
)
from megengine.quantization.quantize import (
disable_fake_quant,
disable_observer,
propagate_qconfig,
)
min_max_fakequant_qconfig = QConfig(
weight_observer=partial(MinMaxObserver, dtype="qint8_narrow"),
act_observer=partial(MinMaxObserver, dtype="qint8"),
weight_fake_quant=partial(FakeQuantize, dtype="qint8_narrow"),
act_fake_quant=partial(FakeQuantize, dtype="qint8"),
)
def gen_inp_scale():
return np.float32(np.random.rand() + 1)
min_val = np.random.randint(-127, 0, size=(2,)).astype("float32")
max_val = np.random.randint(1, 127, size=(2,)).astype("float32")
weight_scale = np.float32(np.max([-min_val[0], max_val[0]]) / 254 * 2)
act_scale = np.float32(np.max([-min_val[1], max_val[1]]) / 255 * 2)
def quant(x, scale):
inp_dtype = dtype.qint8(scale)
return x.astype(inp_dtype)
def fake_quant(x, scale, qmin, qmax):
x = x / scale
x = F.round(x)
x = F.clip(x, qmin, qmax)
x = x * scale
return x
fake_quant_act = partial(fake_quant, qmin=-128, qmax=127)
fake_quant_weight = partial(fake_quant, qmin=-127, qmax=127)
fake_quant_bias = partial(fake_quant, qmin=-(2 ** 31), qmax=2 ** 31 - 1)
def init_qat_net(net):
if net.with_weight:
net.weight_observer.min_val[...] = Tensor(min_val[0])
net.weight_observer.max_val[...] = Tensor(max_val[0])
if net.with_act:
net.act_observer.min_val[...] = Tensor(min_val[1])
net.act_observer.max_val[...] = Tensor(max_val[1])
def test_quant_stub():
normal_net = Float.QuantStub()
normal_net.eval()
qat_from_float = QAT.QuantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.QuantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.QuantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_dequant_stub():
normal_net = Float.DequantStub()
normal_net.eval()
qat_from_float = QAT.DequantStub.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
qat_net = QAT.DequantStub()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.DequantStub.from_qat_module(qat_net)
q_net.eval()
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.scale = inp_scale
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = normal_net(x)
qat = qat_net(x)
q = q_net(quant(x, inp_scale)).numpy()
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("kind", ["cos", "relu", "add", "mul", "fuse_add_relu"])
def test_elemwise(kind):
normal_net = Float.Elemwise(kind)
normal_net.eval()
qat_from_float = QAT.Elemwise.from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
qat_net = QAT.Elemwise(kind)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
q_net = Q.Elemwise.from_qat_module(qat_net)
q_net.eval()
x1_scale = np.float32(np.random.rand() + 1)
x1 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x1 = fake_quant_act(x1, x1_scale)
x1.qparams.scale = x1_scale
x2_scale = np.float32(np.random.rand() + 1)
x2 = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
x2 = fake_quant_act(x2, x2_scale)
x2.qparams.scale = x2_scale
x1_int8 = quant(x1, x1_scale)
x2_int8 = quant(x2, x2_scale)
# test correctness of `Float`, `QAT` and `Quantized`
if kind in ("add", "mul", "fuse_add_relu"):
normal = normal_net(x1, x2)
qat_without_fakequant = qat_from_float(x1, x2)
fake_quant_normal = fake_quant_act(normal_net(x1, x2), act_scale)
qat = qat_net(x1, x2)
q = q_net(x1_int8, x2_int8).numpy() * act_scale
else:
normal = normal_net(x1)
qat_without_fakequant = qat_from_float(x1)
fake_quant_normal = fake_quant_act(normal_net(x1), act_scale)
qat = qat_net(x1)
q = q_net(x1_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal)
np.testing.assert_allclose(q, fake_quant_normal.numpy())
def test_linear():
normal_net = Float.Linear(3, 3, bias=True)
normal_net.eval()
qat_net = QAT.Linear(3, 3, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3)).astype("float32")
bias = np.random.normal(size=(3,)).astype("float32")
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = QAT.Linear.from_float_module(normal_net)
qat_from_float.eval()
disable_fake_quant(qat_from_float)
disable_observer(qat_from_float)
q_net = Q.Linear.from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal)
np.testing.assert_allclose(qat, fake_quant_normal.numpy())
np.testing.assert_allclose(q, fake_quant_normal.numpy())
@pytest.mark.parametrize("module", ["Conv2d", "ConvBn2d", "ConvBnRelu2d"])
def test_conv(module):
normal_net = getattr(Float, module)(3, 3, 3, 1, 1, 1, bias=True)
normal_net.eval()
qat_net = getattr(QAT, module)(3, 3, 3, 1, 1, 1, bias=True)
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
x = mge.tensor(np.random.normal(size=(1, 3, 3, 3)).astype("float32"))
inp_scale = gen_inp_scale()
x = fake_quant_act(x, inp_scale)
x.qparams.update(create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale))
x_int8 = quant(x, inp_scale)
weight = np.random.normal(size=(3, 3, 3, 3)).astype("float32")
bias = np.random.normal(size=(1, 3, 1, 1)).astype("float32")
if module in ("ConvBn2d", "ConvBnRelu2d"):
normal_net.conv.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.conv.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.conv.weight[...] = Parameter(weight)
qat_net.conv.bias[...] = Parameter(bias)
else:
normal_net.weight[...] = fake_quant_weight(weight, weight_scale)
normal_net.bias[...] = fake_quant_bias(bias, inp_scale * weight_scale)
qat_net.weight[...] = Parameter(weight)
qat_net.bias[...] = Parameter(bias)
qat_from_float = getattr(QAT, module).from_float_module(normal_net)
qat_from_float.eval()
disable_observer(qat_from_float)
disable_fake_quant(qat_from_float)
q_net = getattr(Q, module).from_qat_module(qat_net)
q_net.eval()
normal = normal_net(x)
qat_without_fakequant = qat_from_float(x)
fake_quant_normal = fake_quant_act(normal_net(x), act_scale)
qat = qat_net(x)
q = q_net(x_int8).numpy() * act_scale
np.testing.assert_allclose(qat_without_fakequant, normal, atol=1e-5)
np.testing.assert_allclose(qat, fake_quant_normal, atol=act_scale)
np.testing.assert_allclose(q, fake_quant_normal.numpy(), atol=act_scale)
def test_concat():
normal_net = Float.Concat()
normal_net.eval()
qat_net = QAT.Concat()
qat_net.eval()
disable_observer(qat_net)
propagate_qconfig(qat_net, min_max_fakequant_qconfig)
init_qat_net(qat_net)
inps = []
inps_int8 = []
for i in range(3):
inp_scale = gen_inp_scale()
inps.append(mge.tensor(np.random.normal(size=(3, 3)).astype("float32")))
inps[i] = fake_quant_act(inps[i], inp_scale)
inps[i].qparams.update( | create_qparams(QuantMode.SYMMERTIC, "qint8", inp_scale) | megengine.quantization.create_qparams |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import argparse
import importlib
import multiprocessing as mp
import os
import sys
import megengine as mge
import megengine.distributed as dist
from basecore.config import ConfigDict
from loguru import logger
from basecls.engine import ClsTester
from basecls.models import build_model, load_model
from basecls.utils import default_logging, registers, set_nccl_env, set_num_threads, setup_logger
def make_parser() -> argparse.ArgumentParser:
"""Build args parser for testing script.
Returns:
The args parser.
"""
parser = argparse.ArgumentParser()
parser.add_argument("-f", "--file", type=str, help="testing process description file")
parser.add_argument("-w", "--weight_file", default=None, type=str, help="weight file")
return parser
@logger.catch
def worker(args: argparse.Namespace):
"""Worker function for testing script.
Args:
args: args for testing script.
"""
logger.info(f"Init process group for gpu{dist.get_rank()} done")
sys.path.append(os.path.dirname(args.file))
module_name = os.path.splitext(os.path.basename(args.file))[0]
current_network = importlib.import_module(module_name)
cfg = current_network.Cfg()
if cfg.output_dir is None:
cfg.output_dir = f"./logs_{module_name}"
cfg.output_dir = os.path.abspath(cfg.output_dir)
if args.weight_file:
cfg.weights = args.weight_file
else:
cfg.weights = os.path.join(cfg.output_dir, "latest.pkl")
cfg.set_mode("freeze")
if dist.get_rank() == 0 and not os.path.exists(cfg.output_dir):
os.makedirs(cfg.output_dir)
| dist.group_barrier() | megengine.distributed.group_barrier |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import argparse
import importlib
import multiprocessing as mp
import os
import sys
import megengine as mge
import megengine.distributed as dist
from basecore.config import ConfigDict
from loguru import logger
from basecls.engine import ClsTester
from basecls.models import build_model, load_model
from basecls.utils import default_logging, registers, set_nccl_env, set_num_threads, setup_logger
def make_parser() -> argparse.ArgumentParser:
"""Build args parser for testing script.
Returns:
The args parser.
"""
parser = argparse.ArgumentParser()
parser.add_argument("-f", "--file", type=str, help="testing process description file")
parser.add_argument("-w", "--weight_file", default=None, type=str, help="weight file")
return parser
@logger.catch
def worker(args: argparse.Namespace):
"""Worker function for testing script.
Args:
args: args for testing script.
"""
logger.info(f"Init process group for gpu{dist.get_rank()} done")
sys.path.append(os.path.dirname(args.file))
module_name = os.path.splitext(os.path.basename(args.file))[0]
current_network = importlib.import_module(module_name)
cfg = current_network.Cfg()
if cfg.output_dir is None:
cfg.output_dir = f"./logs_{module_name}"
cfg.output_dir = os.path.abspath(cfg.output_dir)
if args.weight_file:
cfg.weights = args.weight_file
else:
cfg.weights = os.path.join(cfg.output_dir, "latest.pkl")
cfg.set_mode("freeze")
if dist.get_rank() == 0 and not os.path.exists(cfg.output_dir):
os.makedirs(cfg.output_dir)
dist.group_barrier()
setup_logger(cfg.output_dir, "test_log.txt", to_loguru=True)
logger.info(f"args: {args}")
if cfg.fastrun:
logger.info("Using fastrun mode...")
mge.functional.debug_param.set_execution_strategy("PROFILE")
tester = build(cfg)
tester.test()
def build(cfg: ConfigDict):
"""Build function for testing script.
Args:
cfg: config for testing.
Returns:
A tester.
"""
model = build_model(cfg)
load_model(model, cfg.weights)
model.eval()
default_logging(cfg, model)
dataloader = registers.dataloaders.get(cfg.data.name).build(cfg, False)
# FIXME: need atomic user_pop, maybe in MegEngine 1.5?
# tester = BaseTester(model, dataloader, AccEvaluator())
return ClsTester(cfg, model, dataloader)
def main():
"""Main function for testing script."""
parser = make_parser()
args = parser.parse_args()
mp.set_start_method("spawn")
set_nccl_env()
set_num_threads()
if not os.path.exists(args.file):
raise ValueError("Description file does not exist")
device_count = | mge.device.get_device_count("gpu") | megengine.device.get_device_count |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import argparse
import importlib
import multiprocessing as mp
import os
import sys
import megengine as mge
import megengine.distributed as dist
from basecore.config import ConfigDict
from loguru import logger
from basecls.engine import ClsTester
from basecls.models import build_model, load_model
from basecls.utils import default_logging, registers, set_nccl_env, set_num_threads, setup_logger
def make_parser() -> argparse.ArgumentParser:
"""Build args parser for testing script.
Returns:
The args parser.
"""
parser = argparse.ArgumentParser()
parser.add_argument("-f", "--file", type=str, help="testing process description file")
parser.add_argument("-w", "--weight_file", default=None, type=str, help="weight file")
return parser
@logger.catch
def worker(args: argparse.Namespace):
"""Worker function for testing script.
Args:
args: args for testing script.
"""
logger.info(f"Init process group for gpu{dist.get_rank()} done")
sys.path.append(os.path.dirname(args.file))
module_name = os.path.splitext(os.path.basename(args.file))[0]
current_network = importlib.import_module(module_name)
cfg = current_network.Cfg()
if cfg.output_dir is None:
cfg.output_dir = f"./logs_{module_name}"
cfg.output_dir = os.path.abspath(cfg.output_dir)
if args.weight_file:
cfg.weights = args.weight_file
else:
cfg.weights = os.path.join(cfg.output_dir, "latest.pkl")
cfg.set_mode("freeze")
if dist.get_rank() == 0 and not os.path.exists(cfg.output_dir):
os.makedirs(cfg.output_dir)
dist.group_barrier()
setup_logger(cfg.output_dir, "test_log.txt", to_loguru=True)
logger.info(f"args: {args}")
if cfg.fastrun:
logger.info("Using fastrun mode...")
| mge.functional.debug_param.set_execution_strategy("PROFILE") | megengine.functional.debug_param.set_execution_strategy |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import argparse
import importlib
import multiprocessing as mp
import os
import sys
import megengine as mge
import megengine.distributed as dist
from basecore.config import ConfigDict
from loguru import logger
from basecls.engine import ClsTester
from basecls.models import build_model, load_model
from basecls.utils import default_logging, registers, set_nccl_env, set_num_threads, setup_logger
def make_parser() -> argparse.ArgumentParser:
"""Build args parser for testing script.
Returns:
The args parser.
"""
parser = argparse.ArgumentParser()
parser.add_argument("-f", "--file", type=str, help="testing process description file")
parser.add_argument("-w", "--weight_file", default=None, type=str, help="weight file")
return parser
@logger.catch
def worker(args: argparse.Namespace):
"""Worker function for testing script.
Args:
args: args for testing script.
"""
logger.info(f"Init process group for gpu{dist.get_rank()} done")
sys.path.append(os.path.dirname(args.file))
module_name = os.path.splitext(os.path.basename(args.file))[0]
current_network = importlib.import_module(module_name)
cfg = current_network.Cfg()
if cfg.output_dir is None:
cfg.output_dir = f"./logs_{module_name}"
cfg.output_dir = os.path.abspath(cfg.output_dir)
if args.weight_file:
cfg.weights = args.weight_file
else:
cfg.weights = os.path.join(cfg.output_dir, "latest.pkl")
cfg.set_mode("freeze")
if | dist.get_rank() | megengine.distributed.get_rank |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import argparse
import importlib
import multiprocessing as mp
import os
import sys
import megengine as mge
import megengine.distributed as dist
from basecore.config import ConfigDict
from loguru import logger
from basecls.engine import ClsTester
from basecls.models import build_model, load_model
from basecls.utils import default_logging, registers, set_nccl_env, set_num_threads, setup_logger
def make_parser() -> argparse.ArgumentParser:
"""Build args parser for testing script.
Returns:
The args parser.
"""
parser = argparse.ArgumentParser()
parser.add_argument("-f", "--file", type=str, help="testing process description file")
parser.add_argument("-w", "--weight_file", default=None, type=str, help="weight file")
return parser
@logger.catch
def worker(args: argparse.Namespace):
"""Worker function for testing script.
Args:
args: args for testing script.
"""
logger.info(f"Init process group for gpu{dist.get_rank()} done")
sys.path.append(os.path.dirname(args.file))
module_name = os.path.splitext(os.path.basename(args.file))[0]
current_network = importlib.import_module(module_name)
cfg = current_network.Cfg()
if cfg.output_dir is None:
cfg.output_dir = f"./logs_{module_name}"
cfg.output_dir = os.path.abspath(cfg.output_dir)
if args.weight_file:
cfg.weights = args.weight_file
else:
cfg.weights = os.path.join(cfg.output_dir, "latest.pkl")
cfg.set_mode("freeze")
if dist.get_rank() == 0 and not os.path.exists(cfg.output_dir):
os.makedirs(cfg.output_dir)
dist.group_barrier()
setup_logger(cfg.output_dir, "test_log.txt", to_loguru=True)
logger.info(f"args: {args}")
if cfg.fastrun:
logger.info("Using fastrun mode...")
mge.functional.debug_param.set_execution_strategy("PROFILE")
tester = build(cfg)
tester.test()
def build(cfg: ConfigDict):
"""Build function for testing script.
Args:
cfg: config for testing.
Returns:
A tester.
"""
model = build_model(cfg)
load_model(model, cfg.weights)
model.eval()
default_logging(cfg, model)
dataloader = registers.dataloaders.get(cfg.data.name).build(cfg, False)
# FIXME: need atomic user_pop, maybe in MegEngine 1.5?
# tester = BaseTester(model, dataloader, AccEvaluator())
return ClsTester(cfg, model, dataloader)
def main():
"""Main function for testing script."""
parser = make_parser()
args = parser.parse_args()
mp.set_start_method("spawn")
set_nccl_env()
set_num_threads()
if not os.path.exists(args.file):
raise ValueError("Description file does not exist")
device_count = mge.device.get_device_count("gpu")
if device_count == 0:
logger.warning("No GPU was found, testing on CPU")
worker(args)
elif device_count > 1:
mp_worker = | dist.launcher(worker) | megengine.distributed.launcher |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import argparse
import importlib
import multiprocessing as mp
import os
import sys
import megengine as mge
import megengine.distributed as dist
from basecore.config import ConfigDict
from loguru import logger
from basecls.engine import ClsTester
from basecls.models import build_model, load_model
from basecls.utils import default_logging, registers, set_nccl_env, set_num_threads, setup_logger
def make_parser() -> argparse.ArgumentParser:
"""Build args parser for testing script.
Returns:
The args parser.
"""
parser = argparse.ArgumentParser()
parser.add_argument("-f", "--file", type=str, help="testing process description file")
parser.add_argument("-w", "--weight_file", default=None, type=str, help="weight file")
return parser
@logger.catch
def worker(args: argparse.Namespace):
"""Worker function for testing script.
Args:
args: args for testing script.
"""
logger.info(f"Init process group for gpu{ | dist.get_rank() | megengine.distributed.get_rank |
#!/usr/bin/env python3
# -*- coding:utf-8 -*-
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import megengine as mge
import megengine.functional as F
import megengine.module as M
__all__ = ["fuse_conv_and_bn", "fuse_model", "replace_module"]
def fuse_conv_and_bn(conv, bn):
# Fuse convolution and batchnorm layers https://tehnokv.com/posts/fusing-batchnorm-and-conv/
fusedconv = M.Conv2d(
conv.in_channels,
conv.out_channels,
kernel_size=conv.kernel_size,
stride=conv.stride,
padding=conv.padding,
groups=conv.groups,
bias=True,
)
# fused_conv.weight = bn.weight / running_var * conv.weight
w_conv = conv.weight.reshape(conv.out_channels, -1)
factor = (bn.weight / F.sqrt(bn.eps + bn.running_var)).reshape(-1)
# diag_factor = diag(factor)
fusedconv.weight = mge.Parameter(
(factor.reshape(-1, 1) * w_conv).reshape(fusedconv.weight.shape)
)
# fused_conv.bias = bn.bias + (conv.bias - running_mean) * bn.weight / runing_var
conv_bias = F.zeros(bn.running_mean.shape) if conv.bias is None else conv.bias
fuse_bias = bn.bias + (conv_bias - bn.running_mean) * factor.reshape(1, -1, 1, 1)
fusedconv.bias = | mge.Parameter(fuse_bias) | megengine.Parameter |
#!/usr/bin/env python3
# -*- coding:utf-8 -*-
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import megengine as mge
import megengine.functional as F
import megengine.module as M
__all__ = ["fuse_conv_and_bn", "fuse_model", "replace_module"]
def fuse_conv_and_bn(conv, bn):
# Fuse convolution and batchnorm layers https://tehnokv.com/posts/fusing-batchnorm-and-conv/
fusedconv = M.Conv2d(
conv.in_channels,
conv.out_channels,
kernel_size=conv.kernel_size,
stride=conv.stride,
padding=conv.padding,
groups=conv.groups,
bias=True,
)
# fused_conv.weight = bn.weight / running_var * conv.weight
w_conv = conv.weight.reshape(conv.out_channels, -1)
factor = (bn.weight / F.sqrt(bn.eps + bn.running_var)).reshape(-1)
# diag_factor = diag(factor)
fusedconv.weight = mge.Parameter(
(factor.reshape(-1, 1) * w_conv).reshape(fusedconv.weight.shape)
)
# fused_conv.bias = bn.bias + (conv.bias - running_mean) * bn.weight / runing_var
conv_bias = | F.zeros(bn.running_mean.shape) | megengine.functional.zeros |
#!/usr/bin/env python3
# -*- coding:utf-8 -*-
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import megengine as mge
import megengine.functional as F
import megengine.module as M
__all__ = ["fuse_conv_and_bn", "fuse_model", "replace_module"]
def fuse_conv_and_bn(conv, bn):
# Fuse convolution and batchnorm layers https://tehnokv.com/posts/fusing-batchnorm-and-conv/
fusedconv = M.Conv2d(
conv.in_channels,
conv.out_channels,
kernel_size=conv.kernel_size,
stride=conv.stride,
padding=conv.padding,
groups=conv.groups,
bias=True,
)
# fused_conv.weight = bn.weight / running_var * conv.weight
w_conv = conv.weight.reshape(conv.out_channels, -1)
factor = (bn.weight / | F.sqrt(bn.eps + bn.running_var) | megengine.functional.sqrt |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import megengine as mge
import megengine.module as M
import numpy as np
import pytest
from basecls.layers import ClsHead, MBV3Head, VGGHead, build_head
@pytest.mark.parametrize("w_in", [4])
@pytest.mark.parametrize(
"head_args",
[
None,
dict(name=None),
dict(name="ClsHead", w_out=8),
dict(name="ClsHead", w_out=8, width=8, norm_name="BN", act_name="relu"),
dict(name="ClsHead", w_out=8, width=8, norm_name="BN", act_name="relu", bias=False),
dict(name="VGGHead", w_out=8, width=8),
dict(name="VGGHead", w_out=8, width=8, dropout_prob=0.5, act_name="relu"),
dict(name="MBV3Head", w_out=8, width=8, w_h=16, norm_name="BN", act_name="relu"),
dict(name="MBV3Head", w_out=8, width=8, w_h=16, norm_name="BN", act_name="relu", se_r=0.25),
dict(
name="MBV3Head",
w_out=8,
width=8,
w_h=16,
norm_name="BN",
act_name="relu",
se_r=0.25,
bias=False,
),
],
)
@pytest.mark.parametrize("norm_name", ["BN"])
@pytest.mark.parametrize("act_name", ["relu"])
def test_build_head(w_in, head_args, norm_name, act_name):
m = build_head(w_in, head_args, norm_name, act_name)
if head_args is None or head_args.get("name") is None:
assert m is None
else:
assert isinstance(m, M.Module)
def test_cls_head():
C = 4
K = 8
x = np.random.rand(2, C, 8, 8).astype("float32")
y = ClsHead(C, K)( | mge.Tensor(x) | megengine.Tensor |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import megengine as mge
import megengine.module as M
import numpy as np
import pytest
from basecls.layers import ClsHead, MBV3Head, VGGHead, build_head
@pytest.mark.parametrize("w_in", [4])
@pytest.mark.parametrize(
"head_args",
[
None,
dict(name=None),
dict(name="ClsHead", w_out=8),
dict(name="ClsHead", w_out=8, width=8, norm_name="BN", act_name="relu"),
dict(name="ClsHead", w_out=8, width=8, norm_name="BN", act_name="relu", bias=False),
dict(name="VGGHead", w_out=8, width=8),
dict(name="VGGHead", w_out=8, width=8, dropout_prob=0.5, act_name="relu"),
dict(name="MBV3Head", w_out=8, width=8, w_h=16, norm_name="BN", act_name="relu"),
dict(name="MBV3Head", w_out=8, width=8, w_h=16, norm_name="BN", act_name="relu", se_r=0.25),
dict(
name="MBV3Head",
w_out=8,
width=8,
w_h=16,
norm_name="BN",
act_name="relu",
se_r=0.25,
bias=False,
),
],
)
@pytest.mark.parametrize("norm_name", ["BN"])
@pytest.mark.parametrize("act_name", ["relu"])
def test_build_head(w_in, head_args, norm_name, act_name):
m = build_head(w_in, head_args, norm_name, act_name)
if head_args is None or head_args.get("name") is None:
assert m is None
else:
assert isinstance(m, M.Module)
def test_cls_head():
C = 4
K = 8
x = np.random.rand(2, C, 8, 8).astype("float32")
y = ClsHead(C, K)(mge.Tensor(x)).numpy()
assert len(y.shape) == 2 and y.shape[1] == K
def test_mbv3_head():
C = 4
K = 8
x = np.random.rand(2, C, 8, 8).astype("float32")
y = MBV3Head(C, K, 8, 16)( | mge.Tensor(x) | megengine.Tensor |
#!/usr/bin/env python3
# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
import megengine as mge
import megengine.module as M
import numpy as np
import pytest
from basecls.layers import ClsHead, MBV3Head, VGGHead, build_head
@pytest.mark.parametrize("w_in", [4])
@pytest.mark.parametrize(
"head_args",
[
None,
dict(name=None),
dict(name="ClsHead", w_out=8),
dict(name="ClsHead", w_out=8, width=8, norm_name="BN", act_name="relu"),
dict(name="ClsHead", w_out=8, width=8, norm_name="BN", act_name="relu", bias=False),
dict(name="VGGHead", w_out=8, width=8),
dict(name="VGGHead", w_out=8, width=8, dropout_prob=0.5, act_name="relu"),
dict(name="MBV3Head", w_out=8, width=8, w_h=16, norm_name="BN", act_name="relu"),
dict(name="MBV3Head", w_out=8, width=8, w_h=16, norm_name="BN", act_name="relu", se_r=0.25),
dict(
name="MBV3Head",
w_out=8,
width=8,
w_h=16,
norm_name="BN",
act_name="relu",
se_r=0.25,
bias=False,
),
],
)
@pytest.mark.parametrize("norm_name", ["BN"])
@pytest.mark.parametrize("act_name", ["relu"])
def test_build_head(w_in, head_args, norm_name, act_name):
m = build_head(w_in, head_args, norm_name, act_name)
if head_args is None or head_args.get("name") is None:
assert m is None
else:
assert isinstance(m, M.Module)
def test_cls_head():
C = 4
K = 8
x = np.random.rand(2, C, 8, 8).astype("float32")
y = ClsHead(C, K)(mge.Tensor(x)).numpy()
assert len(y.shape) == 2 and y.shape[1] == K
def test_mbv3_head():
C = 4
K = 8
x = np.random.rand(2, C, 8, 8).astype("float32")
y = MBV3Head(C, K, 8, 16)(mge.Tensor(x)).numpy()
assert len(y.shape) == 2 and y.shape[1] == K
def test_vgg_head():
C = 4
K = 8
x = np.random.rand(2, C, 8, 8).astype("float32")
y = VGGHead(C, K, 8)( | mge.Tensor(x) | megengine.Tensor |
import os
import os.path as osp
import bisect
import argparse
import multiprocessing as mp
import numpy as np
from tqdm import tqdm
import megengine as mge
from megengine import distributed as dist
from megengine import optimizer as optim
import megengine.autodiff as autodiff
from megengine import jit
import dataset, network
from config import config as cfg
from misc_utils import ensure_dir
import socket
import pdb
ensure_dir(cfg.output_dir)
logger = | mge.get_logger(__name__) | megengine.get_logger |
import os
import os.path as osp
import bisect
import argparse
import multiprocessing as mp
import numpy as np
from tqdm import tqdm
import megengine as mge
from megengine import distributed as dist
from megengine import optimizer as optim
import megengine.autodiff as autodiff
from megengine import jit
import dataset, network
from config import config as cfg
from misc_utils import ensure_dir
import socket
import pdb
ensure_dir(cfg.output_dir)
logger = mge.get_logger(__name__)
log_path = osp.join(cfg.output_dir, 'logger.log')
| mge.set_log_file(log_path, mode='a') | megengine.set_log_file |
import os
import os.path as osp
import bisect
import argparse
import multiprocessing as mp
import numpy as np
from tqdm import tqdm
import megengine as mge
from megengine import distributed as dist
from megengine import optimizer as optim
import megengine.autodiff as autodiff
from megengine import jit
import dataset, network
from config import config as cfg
from misc_utils import ensure_dir
import socket
import pdb
ensure_dir(cfg.output_dir)
logger = mge.get_logger(__name__)
log_path = osp.join(cfg.output_dir, 'logger.log')
mge.set_log_file(log_path, mode='a')
def find_free_port():
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind(("", 0))
port = sock.getsockname()[1]
sock.close()
return port
def allreduce_cb(param, grad, group=dist.WORLD):
return dist.functional.all_reduce_sum(grad, group) / group.size
def train_one_epoch(model, gm, data_iter, opt, max_steps, rank, epoch_id, gpu_num):
# @jit.trace(symbolic=False,)
def propagate():
with gm:
loss_dict = model(model.inputs)
total_loss = sum([loss_dict[key].mean() for key in loss_dict.keys()])
gm.backward(total_loss)
opt.step().clear_grad()
loss_dict['total_loss'] = total_loss
return loss_dict
for step in range(max_steps):
# learing rate
if epoch_id == 0 and step < cfg.warm_iters:
base_lr = (
cfg.basic_lr * gpu_num * cfg.batch_per_gpu
* (cfg.lr_decay_rate ** bisect.bisect_right(cfg.lr_decay_sates, epoch_id)
)
)
lr_factor = (step + 1.0) / cfg.warm_iters
for param_group in opt.param_groups:
param_group['lr'] = (0.33 + 0.67 * lr_factor) * base_lr
mini_batch = next(data_iter)
im_info = mini_batch["im_info"]
image = mini_batch["data"][:, :, :int(im_info[0, 0]), :int(im_info[0, 1])]
model.inputs["image"].set_value(image)
model.inputs["gt_boxes"].set_value(mini_batch["boxes"])
model.inputs["im_info"].set_value(mini_batch["im_info"])
del mini_batch
losses = propagate()
print_str = ' '
for loss_name, loss_value in losses.items():
print_str += ', {}: {:.4f}'.format(loss_name, loss_value.numpy())
if rank == 0:
if step % cfg.log_dump_interval == 0:
logger.info(
"epoch-{}, {}/{}, lr: {:.4f}{}.\n{}".format(
epoch_id,
step,
max_steps,
opt.param_groups[0]["lr"],
print_str,
cfg.workspace,
)
)
def worker(rank, gpu_num, args):
# using sublinear
os.environ["MGB_COMP_GRAPH_OPT"] = "enable_sublinear_memory_opt=1;seq_opt.enable_seq_comp_node_opt=0"
os.environ["MGB_SUBLINEAR_MEMORY_GENETIC_NR_ITER"] = '10'
os.environ['MGB_CUDA_RESERVE_MEMORY'] = '1'
# establish the server if is the master
dist_port = args.port
if rank == 0:
dist.Server(port=dist_port)
if gpu_num> 1:
dist.init_process_group(
master_ip="localhost",
port=dist_port,
world_size=gpu_num,
rank=rank,
device=rank,
)
logger.info("Init process group for gpu%d done", rank)
model = network.Network()
params = model.parameters(requires_grad=True)
model.train()
# Autodiff gradient manager
gm = autodiff.GradManager().attach(
model.parameters(),
callbacks=allreduce_cb,
)
opt = optim.SGD(
params,
lr=cfg.basic_lr * gpu_num * cfg.batch_per_gpu,
momentum=cfg.momentum,
weight_decay=cfg.weight_decay,
)
if cfg.pretrain_weight is not None:
weights = mge.load(cfg.pretrain_weight)
del weights['fc.weight']
del weights['fc.bias']
model.resnet50.load_state_dict(weights)
start_epoch = 0
if args.resume_weights is not None:
assert osp.exists(args.resume_weights)
model_file = args.resume_weights
model_dict = mge.load(model_file)
start_epoch, weights = model_dict['epoch'] + 1, model_dict['state_dict']
model.load_state_dict(weights, strict=False)
logger.info("Prepare dataset")
train_loader = dataset.train_dataset(rank)
logger.info("Training...")
for epoch_id in range(start_epoch, cfg.max_epoch):
for param_group in opt.param_groups:
param_group["lr"] = (
cfg.basic_lr * gpu_num * cfg.batch_per_gpu
* (cfg.lr_decay_rate ** bisect.bisect_right(cfg.lr_decay_sates, epoch_id))
)
max_steps = cfg.nr_images_epoch // (cfg.batch_per_gpu * gpu_num)
train_one_epoch(model, gm, train_loader, opt, max_steps, rank, epoch_id, gpu_num)
if rank == 0:
save_path = osp.join(cfg.model_dir, 'epoch-{}.pkl'.format(epoch_id + 1))
state_dict = model.state_dict()
names = [k for k, _ in state_dict.items()]
for name in names:
if name.startswith('inputs.'):
del state_dict[name]
mge.save(
{"epoch": epoch_id, "state_dict": state_dict}, save_path,
)
logger.info("dump weights to %s", save_path)
def train(args):
# ------------------------ begin training -------------------------- #
valid_nr_dev = | mge.get_device_count("gpu") | megengine.get_device_count |
import os
import os.path as osp
import bisect
import argparse
import multiprocessing as mp
import numpy as np
from tqdm import tqdm
import megengine as mge
from megengine import distributed as dist
from megengine import optimizer as optim
import megengine.autodiff as autodiff
from megengine import jit
import dataset, network
from config import config as cfg
from misc_utils import ensure_dir
import socket
import pdb
ensure_dir(cfg.output_dir)
logger = mge.get_logger(__name__)
log_path = osp.join(cfg.output_dir, 'logger.log')
mge.set_log_file(log_path, mode='a')
def find_free_port():
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind(("", 0))
port = sock.getsockname()[1]
sock.close()
return port
def allreduce_cb(param, grad, group=dist.WORLD):
return | dist.functional.all_reduce_sum(grad, group) | megengine.distributed.functional.all_reduce_sum |
import os
import os.path as osp
import bisect
import argparse
import multiprocessing as mp
import numpy as np
from tqdm import tqdm
import megengine as mge
from megengine import distributed as dist
from megengine import optimizer as optim
import megengine.autodiff as autodiff
from megengine import jit
import dataset, network
from config import config as cfg
from misc_utils import ensure_dir
import socket
import pdb
ensure_dir(cfg.output_dir)
logger = mge.get_logger(__name__)
log_path = osp.join(cfg.output_dir, 'logger.log')
mge.set_log_file(log_path, mode='a')
def find_free_port():
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind(("", 0))
port = sock.getsockname()[1]
sock.close()
return port
def allreduce_cb(param, grad, group=dist.WORLD):
return dist.functional.all_reduce_sum(grad, group) / group.size
def train_one_epoch(model, gm, data_iter, opt, max_steps, rank, epoch_id, gpu_num):
# @jit.trace(symbolic=False,)
def propagate():
with gm:
loss_dict = model(model.inputs)
total_loss = sum([loss_dict[key].mean() for key in loss_dict.keys()])
gm.backward(total_loss)
opt.step().clear_grad()
loss_dict['total_loss'] = total_loss
return loss_dict
for step in range(max_steps):
# learing rate
if epoch_id == 0 and step < cfg.warm_iters:
base_lr = (
cfg.basic_lr * gpu_num * cfg.batch_per_gpu
* (cfg.lr_decay_rate ** bisect.bisect_right(cfg.lr_decay_sates, epoch_id)
)
)
lr_factor = (step + 1.0) / cfg.warm_iters
for param_group in opt.param_groups:
param_group['lr'] = (0.33 + 0.67 * lr_factor) * base_lr
mini_batch = next(data_iter)
im_info = mini_batch["im_info"]
image = mini_batch["data"][:, :, :int(im_info[0, 0]), :int(im_info[0, 1])]
model.inputs["image"].set_value(image)
model.inputs["gt_boxes"].set_value(mini_batch["boxes"])
model.inputs["im_info"].set_value(mini_batch["im_info"])
del mini_batch
losses = propagate()
print_str = ' '
for loss_name, loss_value in losses.items():
print_str += ', {}: {:.4f}'.format(loss_name, loss_value.numpy())
if rank == 0:
if step % cfg.log_dump_interval == 0:
logger.info(
"epoch-{}, {}/{}, lr: {:.4f}{}.\n{}".format(
epoch_id,
step,
max_steps,
opt.param_groups[0]["lr"],
print_str,
cfg.workspace,
)
)
def worker(rank, gpu_num, args):
# using sublinear
os.environ["MGB_COMP_GRAPH_OPT"] = "enable_sublinear_memory_opt=1;seq_opt.enable_seq_comp_node_opt=0"
os.environ["MGB_SUBLINEAR_MEMORY_GENETIC_NR_ITER"] = '10'
os.environ['MGB_CUDA_RESERVE_MEMORY'] = '1'
# establish the server if is the master
dist_port = args.port
if rank == 0:
| dist.Server(port=dist_port) | megengine.distributed.Server |
import os
import os.path as osp
import bisect
import argparse
import multiprocessing as mp
import numpy as np
from tqdm import tqdm
import megengine as mge
from megengine import distributed as dist
from megengine import optimizer as optim
import megengine.autodiff as autodiff
from megengine import jit
import dataset, network
from config import config as cfg
from misc_utils import ensure_dir
import socket
import pdb
ensure_dir(cfg.output_dir)
logger = mge.get_logger(__name__)
log_path = osp.join(cfg.output_dir, 'logger.log')
mge.set_log_file(log_path, mode='a')
def find_free_port():
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind(("", 0))
port = sock.getsockname()[1]
sock.close()
return port
def allreduce_cb(param, grad, group=dist.WORLD):
return dist.functional.all_reduce_sum(grad, group) / group.size
def train_one_epoch(model, gm, data_iter, opt, max_steps, rank, epoch_id, gpu_num):
# @jit.trace(symbolic=False,)
def propagate():
with gm:
loss_dict = model(model.inputs)
total_loss = sum([loss_dict[key].mean() for key in loss_dict.keys()])
gm.backward(total_loss)
opt.step().clear_grad()
loss_dict['total_loss'] = total_loss
return loss_dict
for step in range(max_steps):
# learing rate
if epoch_id == 0 and step < cfg.warm_iters:
base_lr = (
cfg.basic_lr * gpu_num * cfg.batch_per_gpu
* (cfg.lr_decay_rate ** bisect.bisect_right(cfg.lr_decay_sates, epoch_id)
)
)
lr_factor = (step + 1.0) / cfg.warm_iters
for param_group in opt.param_groups:
param_group['lr'] = (0.33 + 0.67 * lr_factor) * base_lr
mini_batch = next(data_iter)
im_info = mini_batch["im_info"]
image = mini_batch["data"][:, :, :int(im_info[0, 0]), :int(im_info[0, 1])]
model.inputs["image"].set_value(image)
model.inputs["gt_boxes"].set_value(mini_batch["boxes"])
model.inputs["im_info"].set_value(mini_batch["im_info"])
del mini_batch
losses = propagate()
print_str = ' '
for loss_name, loss_value in losses.items():
print_str += ', {}: {:.4f}'.format(loss_name, loss_value.numpy())
if rank == 0:
if step % cfg.log_dump_interval == 0:
logger.info(
"epoch-{}, {}/{}, lr: {:.4f}{}.\n{}".format(
epoch_id,
step,
max_steps,
opt.param_groups[0]["lr"],
print_str,
cfg.workspace,
)
)
def worker(rank, gpu_num, args):
# using sublinear
os.environ["MGB_COMP_GRAPH_OPT"] = "enable_sublinear_memory_opt=1;seq_opt.enable_seq_comp_node_opt=0"
os.environ["MGB_SUBLINEAR_MEMORY_GENETIC_NR_ITER"] = '10'
os.environ['MGB_CUDA_RESERVE_MEMORY'] = '1'
# establish the server if is the master
dist_port = args.port
if rank == 0:
dist.Server(port=dist_port)
if gpu_num> 1:
dist.init_process_group(
master_ip="localhost",
port=dist_port,
world_size=gpu_num,
rank=rank,
device=rank,
)
logger.info("Init process group for gpu%d done", rank)
model = network.Network()
params = model.parameters(requires_grad=True)
model.train()
# Autodiff gradient manager
gm = autodiff.GradManager().attach(
model.parameters(),
callbacks=allreduce_cb,
)
opt = optim.SGD(
params,
lr=cfg.basic_lr * gpu_num * cfg.batch_per_gpu,
momentum=cfg.momentum,
weight_decay=cfg.weight_decay,
)
if cfg.pretrain_weight is not None:
weights = | mge.load(cfg.pretrain_weight) | megengine.load |
import os
import os.path as osp
import bisect
import argparse
import multiprocessing as mp
import numpy as np
from tqdm import tqdm
import megengine as mge
from megengine import distributed as dist
from megengine import optimizer as optim
import megengine.autodiff as autodiff
from megengine import jit
import dataset, network
from config import config as cfg
from misc_utils import ensure_dir
import socket
import pdb
ensure_dir(cfg.output_dir)
logger = mge.get_logger(__name__)
log_path = osp.join(cfg.output_dir, 'logger.log')
mge.set_log_file(log_path, mode='a')
def find_free_port():
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind(("", 0))
port = sock.getsockname()[1]
sock.close()
return port
def allreduce_cb(param, grad, group=dist.WORLD):
return dist.functional.all_reduce_sum(grad, group) / group.size
def train_one_epoch(model, gm, data_iter, opt, max_steps, rank, epoch_id, gpu_num):
# @jit.trace(symbolic=False,)
def propagate():
with gm:
loss_dict = model(model.inputs)
total_loss = sum([loss_dict[key].mean() for key in loss_dict.keys()])
gm.backward(total_loss)
opt.step().clear_grad()
loss_dict['total_loss'] = total_loss
return loss_dict
for step in range(max_steps):
# learing rate
if epoch_id == 0 and step < cfg.warm_iters:
base_lr = (
cfg.basic_lr * gpu_num * cfg.batch_per_gpu
* (cfg.lr_decay_rate ** bisect.bisect_right(cfg.lr_decay_sates, epoch_id)
)
)
lr_factor = (step + 1.0) / cfg.warm_iters
for param_group in opt.param_groups:
param_group['lr'] = (0.33 + 0.67 * lr_factor) * base_lr
mini_batch = next(data_iter)
im_info = mini_batch["im_info"]
image = mini_batch["data"][:, :, :int(im_info[0, 0]), :int(im_info[0, 1])]
model.inputs["image"].set_value(image)
model.inputs["gt_boxes"].set_value(mini_batch["boxes"])
model.inputs["im_info"].set_value(mini_batch["im_info"])
del mini_batch
losses = propagate()
print_str = ' '
for loss_name, loss_value in losses.items():
print_str += ', {}: {:.4f}'.format(loss_name, loss_value.numpy())
if rank == 0:
if step % cfg.log_dump_interval == 0:
logger.info(
"epoch-{}, {}/{}, lr: {:.4f}{}.\n{}".format(
epoch_id,
step,
max_steps,
opt.param_groups[0]["lr"],
print_str,
cfg.workspace,
)
)
def worker(rank, gpu_num, args):
# using sublinear
os.environ["MGB_COMP_GRAPH_OPT"] = "enable_sublinear_memory_opt=1;seq_opt.enable_seq_comp_node_opt=0"
os.environ["MGB_SUBLINEAR_MEMORY_GENETIC_NR_ITER"] = '10'
os.environ['MGB_CUDA_RESERVE_MEMORY'] = '1'
# establish the server if is the master
dist_port = args.port
if rank == 0:
dist.Server(port=dist_port)
if gpu_num> 1:
dist.init_process_group(
master_ip="localhost",
port=dist_port,
world_size=gpu_num,
rank=rank,
device=rank,
)
logger.info("Init process group for gpu%d done", rank)
model = network.Network()
params = model.parameters(requires_grad=True)
model.train()
# Autodiff gradient manager
gm = autodiff.GradManager().attach(
model.parameters(),
callbacks=allreduce_cb,
)
opt = optim.SGD(
params,
lr=cfg.basic_lr * gpu_num * cfg.batch_per_gpu,
momentum=cfg.momentum,
weight_decay=cfg.weight_decay,
)
if cfg.pretrain_weight is not None:
weights = mge.load(cfg.pretrain_weight)
del weights['fc.weight']
del weights['fc.bias']
model.resnet50.load_state_dict(weights)
start_epoch = 0
if args.resume_weights is not None:
assert osp.exists(args.resume_weights)
model_file = args.resume_weights
model_dict = | mge.load(model_file) | megengine.load |
import os
import os.path as osp
import bisect
import argparse
import multiprocessing as mp
import numpy as np
from tqdm import tqdm
import megengine as mge
from megengine import distributed as dist
from megengine import optimizer as optim
import megengine.autodiff as autodiff
from megengine import jit
import dataset, network
from config import config as cfg
from misc_utils import ensure_dir
import socket
import pdb
ensure_dir(cfg.output_dir)
logger = mge.get_logger(__name__)
log_path = osp.join(cfg.output_dir, 'logger.log')
mge.set_log_file(log_path, mode='a')
def find_free_port():
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.bind(("", 0))
port = sock.getsockname()[1]
sock.close()
return port
def allreduce_cb(param, grad, group=dist.WORLD):
return dist.functional.all_reduce_sum(grad, group) / group.size
def train_one_epoch(model, gm, data_iter, opt, max_steps, rank, epoch_id, gpu_num):
# @jit.trace(symbolic=False,)
def propagate():
with gm:
loss_dict = model(model.inputs)
total_loss = sum([loss_dict[key].mean() for key in loss_dict.keys()])
gm.backward(total_loss)
opt.step().clear_grad()
loss_dict['total_loss'] = total_loss
return loss_dict
for step in range(max_steps):
# learing rate
if epoch_id == 0 and step < cfg.warm_iters:
base_lr = (
cfg.basic_lr * gpu_num * cfg.batch_per_gpu
* (cfg.lr_decay_rate ** bisect.bisect_right(cfg.lr_decay_sates, epoch_id)
)
)
lr_factor = (step + 1.0) / cfg.warm_iters
for param_group in opt.param_groups:
param_group['lr'] = (0.33 + 0.67 * lr_factor) * base_lr
mini_batch = next(data_iter)
im_info = mini_batch["im_info"]
image = mini_batch["data"][:, :, :int(im_info[0, 0]), :int(im_info[0, 1])]
model.inputs["image"].set_value(image)
model.inputs["gt_boxes"].set_value(mini_batch["boxes"])
model.inputs["im_info"].set_value(mini_batch["im_info"])
del mini_batch
losses = propagate()
print_str = ' '
for loss_name, loss_value in losses.items():
print_str += ', {}: {:.4f}'.format(loss_name, loss_value.numpy())
if rank == 0:
if step % cfg.log_dump_interval == 0:
logger.info(
"epoch-{}, {}/{}, lr: {:.4f}{}.\n{}".format(
epoch_id,
step,
max_steps,
opt.param_groups[0]["lr"],
print_str,
cfg.workspace,
)
)
def worker(rank, gpu_num, args):
# using sublinear
os.environ["MGB_COMP_GRAPH_OPT"] = "enable_sublinear_memory_opt=1;seq_opt.enable_seq_comp_node_opt=0"
os.environ["MGB_SUBLINEAR_MEMORY_GENETIC_NR_ITER"] = '10'
os.environ['MGB_CUDA_RESERVE_MEMORY'] = '1'
# establish the server if is the master
dist_port = args.port
if rank == 0:
dist.Server(port=dist_port)
if gpu_num> 1:
dist.init_process_group(
master_ip="localhost",
port=dist_port,
world_size=gpu_num,
rank=rank,
device=rank,
)
logger.info("Init process group for gpu%d done", rank)
model = network.Network()
params = model.parameters(requires_grad=True)
model.train()
# Autodiff gradient manager
gm = | autodiff.GradManager() | megengine.autodiff.GradManager |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = | mge.get_logger(__name__) | megengine.get_logger |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = | dist.get_rank() | megengine.distributed.get_rank |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = | mge.load(args.checkpoint) | megengine.load |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = | data.dataset.ImageNet(args.data, train=False) | megengine.data.dataset.ImageNet |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), T.ToMode("CHW")]
),
num_workers=args.workers,
)
# calibration
model.fc.disable_quantize()
model = | quantize_qat(model, qconfig=Q.calibration_qconfig) | megengine.quantization.quantize.quantize_qat |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), T.ToMode("CHW")]
),
num_workers=args.workers,
)
# calibration
model.fc.disable_quantize()
model = quantize_qat(model, qconfig=Q.calibration_qconfig)
# calculate scale
def calculate_scale(image, label):
model.eval()
enable_observer(model)
logits = model(image)
loss = F.loss.cross_entropy(logits, label, label_smooth=0.1)
acc1, acc5 = F.topk_accuracy(logits, label, (1, 5))
if dist.is_distributed(): # all_reduce_mean
loss = dist.functional.all_reduce_sum(loss) / dist.get_world_size()
acc1 = dist.functional.all_reduce_sum(acc1) / dist.get_world_size()
acc5 = dist.functional.all_reduce_sum(acc5) / dist.get_world_size()
return loss, acc1, acc5
infer(calculate_scale, valid_queue, args)
# quantized
model = | quantize(model) | megengine.quantization.quantize.quantize |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
| dist.helper.get_device_count_by_fork("gpu") | megengine.distributed.helper.get_device_count_by_fork |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = | dist.launcher(worker) | megengine.distributed.launcher |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), T.ToMode("CHW")]
),
num_workers=args.workers,
)
# calibration
model.fc.disable_quantize()
model = quantize_qat(model, qconfig=Q.calibration_qconfig)
# calculate scale
def calculate_scale(image, label):
model.eval()
| enable_observer(model) | megengine.quantization.quantize.enable_observer |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), T.ToMode("CHW")]
),
num_workers=args.workers,
)
# calibration
model.fc.disable_quantize()
model = quantize_qat(model, qconfig=Q.calibration_qconfig)
# calculate scale
def calculate_scale(image, label):
model.eval()
enable_observer(model)
logits = model(image)
loss = | F.loss.cross_entropy(logits, label, label_smooth=0.1) | megengine.functional.loss.cross_entropy |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), T.ToMode("CHW")]
),
num_workers=args.workers,
)
# calibration
model.fc.disable_quantize()
model = quantize_qat(model, qconfig=Q.calibration_qconfig)
# calculate scale
def calculate_scale(image, label):
model.eval()
enable_observer(model)
logits = model(image)
loss = F.loss.cross_entropy(logits, label, label_smooth=0.1)
acc1, acc5 = | F.topk_accuracy(logits, label, (1, 5)) | megengine.functional.topk_accuracy |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), T.ToMode("CHW")]
),
num_workers=args.workers,
)
# calibration
model.fc.disable_quantize()
model = quantize_qat(model, qconfig=Q.calibration_qconfig)
# calculate scale
def calculate_scale(image, label):
model.eval()
enable_observer(model)
logits = model(image)
loss = F.loss.cross_entropy(logits, label, label_smooth=0.1)
acc1, acc5 = F.topk_accuracy(logits, label, (1, 5))
if | dist.is_distributed() | megengine.distributed.is_distributed |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), T.ToMode("CHW")]
),
num_workers=args.workers,
)
# calibration
model.fc.disable_quantize()
model = quantize_qat(model, qconfig=Q.calibration_qconfig)
# calculate scale
def calculate_scale(image, label):
model.eval()
enable_observer(model)
logits = model(image)
loss = F.loss.cross_entropy(logits, label, label_smooth=0.1)
acc1, acc5 = F.topk_accuracy(logits, label, (1, 5))
if dist.is_distributed(): # all_reduce_mean
loss = dist.functional.all_reduce_sum(loss) / dist.get_world_size()
acc1 = dist.functional.all_reduce_sum(acc1) / dist.get_world_size()
acc5 = dist.functional.all_reduce_sum(acc5) / dist.get_world_size()
return loss, acc1, acc5
infer(calculate_scale, valid_queue, args)
# quantized
model = quantize(model)
# eval quantized model
def eval_func(image, label):
model.eval()
logits = model(image)
loss = | F.loss.cross_entropy(logits, label, label_smooth=0.1) | megengine.functional.loss.cross_entropy |
# -*- coding: utf-8 -*-
# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
#
# Copyright (c) 2014-2020 Megvii Inc. All rights reserved.
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
"""Finetune a pretrained fp32 with int8 post train quantization(calibration)"""
import argparse
import collections
import numbers
import os
import time
# pylint: disable=import-error
import models
import megengine as mge
import megengine.data as data
import megengine.data.transform as T
import megengine.distributed as dist
import megengine.functional as F
import megengine.quantization as Q
from megengine.quantization.quantize import enable_observer, quantize, quantize_qat
logger = mge.get_logger(__name__)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("-a", "--arch", default="resnet18", type=str)
parser.add_argument("-d", "--data", default=None, type=str)
parser.add_argument("-s", "--save", default="/data/models", type=str)
parser.add_argument(
"-c",
"--checkpoint",
default=None,
type=str,
help="pretrained model to finetune",
)
parser.add_argument("-n", "--ngpus", default=None, type=int)
parser.add_argument("-w", "--workers", default=4, type=int)
parser.add_argument("--report-freq", default=50, type=int)
args = parser.parse_args()
world_size = (
dist.helper.get_device_count_by_fork("gpu")
if args.ngpus is None
else args.ngpus
)
world_size = 1 if world_size == 0 else world_size
if world_size != 1:
logger.warning(
"Calibration only supports single GPU now, %d provided", world_size
)
proc_func = dist.launcher(worker) if world_size > 1 else worker
proc_func(world_size, args)
def get_parameters(model, cfg):
if isinstance(cfg.WEIGHT_DECAY, numbers.Number):
return {
"params": model.parameters(requires_grad=True),
"weight_decay": cfg.WEIGHT_DECAY,
}
groups = collections.defaultdict(list) # weight_decay -> List[param]
for pname, p in model.named_parameters(requires_grad=True):
wd = cfg.WEIGHT_DECAY(pname, p)
groups[wd].append(p)
groups = [
{"params": params, "weight_decay": wd} for wd, params in groups.items()
] # List[{param, weight_decay}]
return groups
def worker(world_size, args):
# pylint: disable=too-many-statements
rank = dist.get_rank()
if world_size > 1:
# Initialize distributed process group
logger.info("init distributed process group {} / {}".format(rank, world_size))
save_dir = os.path.join(args.save, args.arch + "." + "calibration")
if not os.path.exists(save_dir):
os.makedirs(save_dir, exist_ok=True)
mge.set_log_file(os.path.join(save_dir, "log.txt"))
model = models.__dict__[args.arch]()
# load calibration model
assert args.checkpoint
logger.info("Load pretrained weights from %s", args.checkpoint)
ckpt = mge.load(args.checkpoint)
ckpt = ckpt["state_dict"] if "state_dict" in ckpt else ckpt
model.load_state_dict(ckpt, strict=False)
# Build valid datasets
valid_dataset = data.dataset.ImageNet(args.data, train=False)
valid_sampler = data.SequentialSampler(
valid_dataset, batch_size=100, drop_last=False
)
valid_queue = data.DataLoader(
valid_dataset,
sampler=valid_sampler,
transform=T.Compose(
[T.Resize(256), T.CenterCrop(224), T.Normalize(mean=128), T.ToMode("CHW")]
),
num_workers=args.workers,
)
# calibration
model.fc.disable_quantize()
model = quantize_qat(model, qconfig=Q.calibration_qconfig)
# calculate scale
def calculate_scale(image, label):
model.eval()
enable_observer(model)
logits = model(image)
loss = F.loss.cross_entropy(logits, label, label_smooth=0.1)
acc1, acc5 = F.topk_accuracy(logits, label, (1, 5))
if dist.is_distributed(): # all_reduce_mean
loss = dist.functional.all_reduce_sum(loss) / dist.get_world_size()
acc1 = dist.functional.all_reduce_sum(acc1) / dist.get_world_size()
acc5 = dist.functional.all_reduce_sum(acc5) / dist.get_world_size()
return loss, acc1, acc5
infer(calculate_scale, valid_queue, args)
# quantized
model = quantize(model)
# eval quantized model
def eval_func(image, label):
model.eval()
logits = model(image)
loss = F.loss.cross_entropy(logits, label, label_smooth=0.1)
acc1, acc5 = | F.topk_accuracy(logits, label, (1, 5)) | megengine.functional.topk_accuracy |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.