Commit
·
615d6d0
1
Parent(s):
e8bf385
FROM nvcr.io/nvidia/pytorch:20.06-py3
Browse files- models/{onnx_export.py → export.py} +29 -17
- models/torchscript_export.py +0 -38
models/{onnx_export.py → export.py}
RENAMED
@@ -1,7 +1,7 @@
|
|
1 |
-
"""Exports a
|
2 |
|
3 |
Usage:
|
4 |
-
$ export PYTHONPATH="$PWD" && python models/
|
5 |
"""
|
6 |
|
7 |
import argparse
|
@@ -17,27 +17,39 @@ if __name__ == '__main__':
|
|
17 |
parser.add_argument('--img-size', nargs='+', type=int, default=[640, 640], help='image size')
|
18 |
parser.add_argument('--batch-size', type=int, default=1, help='batch size')
|
19 |
opt = parser.parse_args()
|
20 |
-
opt.img_size *= 2 if len(opt.img_size) == 1 else 1
|
21 |
print(opt)
|
22 |
|
23 |
-
#
|
24 |
-
f = opt.weights.replace('.pt', '.onnx') # onnx filename
|
25 |
img = torch.zeros((opt.batch_size, 3, *opt.img_size)) # image size, (1, 3, 320, 192) iDetection
|
26 |
|
27 |
-
# Load
|
28 |
google_utils.attempt_download(opt.weights)
|
29 |
model = torch.load(opt.weights, map_location=torch.device('cpu'))['model'].float()
|
30 |
model.eval()
|
31 |
-
model.fuse()
|
32 |
-
|
33 |
-
# Export to onnx
|
34 |
model.model[-1].export = True # set Detect() layer export=True
|
35 |
_ = model(img) # dry run
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
"""Exports a YOLOv5 *.pt model to *.onnx and *.torchscript formats
|
2 |
|
3 |
Usage:
|
4 |
+
$ export PYTHONPATH="$PWD" && python models/export.py --weights ./weights/yolov5s.pt --img 640 --batch 1
|
5 |
"""
|
6 |
|
7 |
import argparse
|
|
|
17 |
parser.add_argument('--img-size', nargs='+', type=int, default=[640, 640], help='image size')
|
18 |
parser.add_argument('--batch-size', type=int, default=1, help='batch size')
|
19 |
opt = parser.parse_args()
|
20 |
+
opt.img_size *= 2 if len(opt.img_size) == 1 else 1 # expand
|
21 |
print(opt)
|
22 |
|
23 |
+
# Input
|
|
|
24 |
img = torch.zeros((opt.batch_size, 3, *opt.img_size)) # image size, (1, 3, 320, 192) iDetection
|
25 |
|
26 |
+
# Load PyTorch model
|
27 |
google_utils.attempt_download(opt.weights)
|
28 |
model = torch.load(opt.weights, map_location=torch.device('cpu'))['model'].float()
|
29 |
model.eval()
|
|
|
|
|
|
|
30 |
model.model[-1].export = True # set Detect() layer export=True
|
31 |
_ = model(img) # dry run
|
32 |
+
|
33 |
+
# Export to torchscript
|
34 |
+
try:
|
35 |
+
f = opt.weights.replace('.pt', '.torchscript') # filename
|
36 |
+
ts = torch.jit.trace(model, img)
|
37 |
+
ts.save(f)
|
38 |
+
print('Torchscript export success, saved as %s' % f)
|
39 |
+
except:
|
40 |
+
print('Torchscript export failed.')
|
41 |
+
|
42 |
+
# Export to ONNX
|
43 |
+
try:
|
44 |
+
f = opt.weights.replace('.pt', '.onnx') # filename
|
45 |
+
model.fuse() # only for ONNX
|
46 |
+
torch.onnx.export(model, img, f, verbose=False, opset_version=11, input_names=['images'],
|
47 |
+
output_names=['output']) # output_names=['classes', 'boxes']
|
48 |
+
|
49 |
+
# Checks
|
50 |
+
onnx_model = onnx.load(f) # load onnx model
|
51 |
+
onnx.checker.check_model(onnx_model) # check onnx model
|
52 |
+
print(onnx.helper.printable_graph(onnx_model.graph)) # print a human readable representation of the graph
|
53 |
+
print('ONNX export success, saved as %s\nView with https://github.com/lutzroeder/netron' % f)
|
54 |
+
except:
|
55 |
+
print('ONNX export failed.')
|
models/torchscript_export.py
DELETED
@@ -1,38 +0,0 @@
|
|
1 |
-
"""Exports a pytorch *.pt model to *.torchscript format
|
2 |
-
|
3 |
-
Usage:
|
4 |
-
$ export PYTHONPATH="$PWD" && python models/torchscript_export.py --weights ./weights/yolov5s.pt --img 640 --batch 1
|
5 |
-
"""
|
6 |
-
|
7 |
-
import argparse
|
8 |
-
|
9 |
-
|
10 |
-
from models.common import *
|
11 |
-
from utils import google_utils
|
12 |
-
|
13 |
-
if __name__ == '__main__':
|
14 |
-
parser = argparse.ArgumentParser()
|
15 |
-
parser.add_argument('--weights', type=str, default='./yolov5s.pt', help='weights path')
|
16 |
-
parser.add_argument('--img-size', nargs='+', type=int, default=[640, 640], help='image size')
|
17 |
-
parser.add_argument('--batch-size', type=int, default=1, help='batch size')
|
18 |
-
opt = parser.parse_args()
|
19 |
-
print(opt)
|
20 |
-
|
21 |
-
# Parameters
|
22 |
-
f = opt.weights.replace('.pt', '.torchscript') # onnx filename
|
23 |
-
img = torch.zeros((opt.batch_size, 3, *opt.img_size)) # image size, (1, 3, 320, 192) iDetection
|
24 |
-
|
25 |
-
# Load pytorch model
|
26 |
-
google_utils.attempt_download(opt.weights)
|
27 |
-
model = torch.load(opt.weights, map_location=torch.device('cpu'))['model'].float()
|
28 |
-
model.eval()
|
29 |
-
|
30 |
-
# Don't fuse layers, it won't work with torchscript exports
|
31 |
-
#model.fuse()
|
32 |
-
|
33 |
-
# Export to jit/torchscript
|
34 |
-
model.model[-1].export = True # set Detect() layer export=True
|
35 |
-
_ = model(img) # dry run
|
36 |
-
|
37 |
-
traced_script_module = torch.jit.trace(model, img)
|
38 |
-
traced_script_module.save(f)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|