Spaces:
Runtime error
Runtime error
File size: 7,846 Bytes
3094730 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 |
# Copyright (c) OpenMMLab. All rights reserved.
from typing import List
import torch.nn as nn
from mmcv.cnn import ConvModule
from mmdet.utils import ConfigType, OptMultiConfig
from mmyolo.registry import MODELS
from ..layers import MaxPoolAndStrideConvBlock, RepVGGBlock, SPPFCSPBlock
from .base_yolo_neck import BaseYOLONeck
@MODELS.register_module()
class YOLOv7PAFPN(BaseYOLONeck):
"""Path Aggregation Network used in YOLOv7.
Args:
in_channels (List[int]): Number of input channels per scale.
out_channels (int): Number of output channels (used at each scale).
block_cfg (dict): Config dict for block.
deepen_factor (float): Depth multiplier, multiply number of
blocks in CSP layer by this amount. Defaults to 1.0.
widen_factor (float): Width multiplier, multiply number of
channels in each layer by this amount. Defaults to 1.0.
spp_expand_ratio (float): Expand ratio of SPPCSPBlock.
Defaults to 0.5.
is_tiny_version (bool): Is tiny version of neck. If True,
it means it is a yolov7 tiny model. Defaults to False.
use_maxpool_in_downsample (bool): Whether maxpooling is
used in downsample layers. Defaults to True.
use_in_channels_in_downsample (bool): MaxPoolAndStrideConvBlock
module input parameters. Defaults to False.
use_repconv_outs (bool): Whether to use `repconv` in the output
layer. Defaults to True.
upsample_feats_cat_first (bool): Whether the output features are
concat first after upsampling in the topdown module.
Defaults to True. Currently only YOLOv7 is false.
freeze_all(bool): Whether to freeze the model. Defaults to False.
norm_cfg (dict): Config dict for normalization layer.
Defaults to dict(type='BN', momentum=0.03, eps=0.001).
act_cfg (dict): Config dict for activation layer.
Defaults to dict(type='SiLU', inplace=True).
init_cfg (dict or list[dict], optional): Initialization config dict.
Defaults to None.
"""
def __init__(self,
in_channels: List[int],
out_channels: List[int],
block_cfg: dict = dict(
type='ELANBlock',
middle_ratio=0.5,
block_ratio=0.25,
num_blocks=4,
num_convs_in_block=1),
deepen_factor: float = 1.0,
widen_factor: float = 1.0,
spp_expand_ratio: float = 0.5,
is_tiny_version: bool = False,
use_maxpool_in_downsample: bool = True,
use_in_channels_in_downsample: bool = False,
use_repconv_outs: bool = True,
upsample_feats_cat_first: bool = False,
freeze_all: bool = False,
norm_cfg: ConfigType = dict(
type='BN', momentum=0.03, eps=0.001),
act_cfg: ConfigType = dict(type='SiLU', inplace=True),
init_cfg: OptMultiConfig = None):
self.is_tiny_version = is_tiny_version
self.use_maxpool_in_downsample = use_maxpool_in_downsample
self.use_in_channels_in_downsample = use_in_channels_in_downsample
self.spp_expand_ratio = spp_expand_ratio
self.use_repconv_outs = use_repconv_outs
self.block_cfg = block_cfg
self.block_cfg.setdefault('norm_cfg', norm_cfg)
self.block_cfg.setdefault('act_cfg', act_cfg)
super().__init__(
in_channels=[
int(channel * widen_factor) for channel in in_channels
],
out_channels=[
int(channel * widen_factor) for channel in out_channels
],
deepen_factor=deepen_factor,
widen_factor=widen_factor,
upsample_feats_cat_first=upsample_feats_cat_first,
freeze_all=freeze_all,
norm_cfg=norm_cfg,
act_cfg=act_cfg,
init_cfg=init_cfg)
def build_reduce_layer(self, idx: int) -> nn.Module:
"""build reduce layer.
Args:
idx (int): layer idx.
Returns:
nn.Module: The reduce layer.
"""
if idx == len(self.in_channels) - 1:
layer = SPPFCSPBlock(
self.in_channels[idx],
self.out_channels[idx],
expand_ratio=self.spp_expand_ratio,
is_tiny_version=self.is_tiny_version,
kernel_sizes=5,
norm_cfg=self.norm_cfg,
act_cfg=self.act_cfg)
else:
layer = ConvModule(
self.in_channels[idx],
self.out_channels[idx],
1,
norm_cfg=self.norm_cfg,
act_cfg=self.act_cfg)
return layer
def build_upsample_layer(self, idx: int) -> nn.Module:
"""build upsample layer."""
return nn.Sequential(
ConvModule(
self.out_channels[idx],
self.out_channels[idx - 1],
1,
norm_cfg=self.norm_cfg,
act_cfg=self.act_cfg),
nn.Upsample(scale_factor=2, mode='nearest'))
def build_top_down_layer(self, idx: int) -> nn.Module:
"""build top down layer.
Args:
idx (int): layer idx.
Returns:
nn.Module: The top down layer.
"""
block_cfg = self.block_cfg.copy()
block_cfg['in_channels'] = self.out_channels[idx - 1] * 2
block_cfg['out_channels'] = self.out_channels[idx - 1]
return MODELS.build(block_cfg)
def build_downsample_layer(self, idx: int) -> nn.Module:
"""build downsample layer.
Args:
idx (int): layer idx.
Returns:
nn.Module: The downsample layer.
"""
if self.use_maxpool_in_downsample and not self.is_tiny_version:
return MaxPoolAndStrideConvBlock(
self.out_channels[idx],
self.out_channels[idx + 1],
use_in_channels_of_middle=self.use_in_channels_in_downsample,
norm_cfg=self.norm_cfg,
act_cfg=self.act_cfg)
else:
return ConvModule(
self.out_channels[idx],
self.out_channels[idx + 1],
3,
stride=2,
padding=1,
norm_cfg=self.norm_cfg,
act_cfg=self.act_cfg)
def build_bottom_up_layer(self, idx: int) -> nn.Module:
"""build bottom up layer.
Args:
idx (int): layer idx.
Returns:
nn.Module: The bottom up layer.
"""
block_cfg = self.block_cfg.copy()
block_cfg['in_channels'] = self.out_channels[idx + 1] * 2
block_cfg['out_channels'] = self.out_channels[idx + 1]
return MODELS.build(block_cfg)
def build_out_layer(self, idx: int) -> nn.Module:
"""build out layer.
Args:
idx (int): layer idx.
Returns:
nn.Module: The out layer.
"""
if len(self.in_channels) == 4:
# P6
return nn.Identity()
out_channels = self.out_channels[idx] * 2
if self.use_repconv_outs:
return RepVGGBlock(
self.out_channels[idx],
out_channels,
3,
norm_cfg=self.norm_cfg,
act_cfg=self.act_cfg)
else:
return ConvModule(
self.out_channels[idx],
out_channels,
3,
padding=1,
norm_cfg=self.norm_cfg,
act_cfg=self.act_cfg)
|