Merge original yolov5 master branch 'origin' into spaces main
60fa02e
xiang-wuu
commited on
Commit History
initial commit
5567ee6
Xiang Shin Wuu
commited on
Revert protobuf<=3.20.1 (#8742)
3e85863
unverified
fix: broken ``is_docker`` check (#8711)
0b5ac22
unverified
Revert "Fix BGR->RGB Bug in albumentations #8641" (#8727)
c775a29
unverified
coremltools>=5.2 for CoreML export (#8725)
d5116bb
unverified
Fix BGR->RGB Bug in albumentations #8641 (#8695)
2e1291f
unverified
New `HUBDatasetStats()` class (#8716)
b367860
unverified
Update dataloaders.py (#8714)
a6f197a
unverified
Avoid cv2 window init code on Windows (#8712)
7215a0f
unverified
Remove `else:` from load_image() (#8692)
0ab303b
unverified
Faster crop saving (#8696)
916bdb1
unverified
Move .dockerignore to root (#8690)
b510957
unverified
Set `torch.cuda.manual_seed_all()` for DDP (#8688)
7f7bd6f
unverified
Add generator and worker seed (#8602)
1c5e92a
unverified
Update hubconf.py to reset logging level to INFO (#8680)
b92430a
unverified
Update warning emojis (#8678)
b17629e
unverified
Update hubconf.py to reset LOGGER.level after load (#8674)
38721de
unverified
Use contextlib's suppress method to silence an error (#8668)
4c1784b
unverified
Check TensorBoard logger before adding graph (#8664)
602d7ff
unverified
Rüdiger Busche
commited on
Upgrade onnxsim to v0.4.1 (#8632)
92e47b8
unverified
Fix `OMP_NUM_THREADS=1` for macOS (#8624)
fbe67e4
unverified
assert torch!=1.12.0 for DDP training (#8621)
9cf5fd5
unverified
Refactor optimizer initialization (#8607)
51fb467
unverified
FROM nvcr.io/nvidia/pytorch:22.06-py3 (#8600)
2430578
unverified
Link fuse() to AutoShape() for Hub models (#8599)
a34b376
unverified
Add `pip install wheel` to avoid legacy `setup.py install` (#8597)
6e86af3
unverified
Update `tqdm>=4.64.0` and `thop>=0.1.1` (#8596)
5c45a4b
unverified
Update `scipy>=1.7.3` (#8595)
cf28dda
unverified
Explicitly set `weight_decay` value (#8592)
7204c1c
unverified
Update requirements.txt comment spacing (#8562)
72a81e7
unverified
Assert `--optimize` not used with cuda device (#8569)
f4b0568
unverified
Raise error on suffix-less model path (#8561)
f8722b4
unverified
Add `thop>=0.1.0` (#8558)
574ceed
unverified
Add TensorRT dependencies (#8553)
fbd3020
unverified
曾逸夫（Zeng Yifu）
commited on
Update metrics.py with IoU protected divides (#8550)
0414637
unverified
CIoU protected divides (#8546)
a84cd02
unverified
val.py `assert ncm == nc` fix (#8545)
7dafd1c
unverified
Fix `LoadImages()` with dataset YAML lists (#8517)
526e650
unverified
XML export `--half` fix (#8522)
c215e87
unverified
Add `--half` arguments to export.py Usage examples (#8516)
63ba0cb
unverified
Properly expose `batch_size` from OpenVINO similarly to TensorRT (#8514)
be42a24
unverified
Simplify benchmarks.py assertions (#8515)
f17444a
unverified
Add `--hard-fail` argument to benchmarks for CI errors (#8513)
3e54651
unverified
Fix AP calculation bug #8464 (#8484)
39d7a93
unverified
Avoid FP64 ops for MPS support in train.py (#8511)
dd28df9
unverified
Revert "Expose OpenVINO `batch_size` similarly to TensorRT" (#8510)
9d7bc06
unverified
Training reproducibility improvements (#8213)
27d831b
unverified
Update tutorial.ipynb (#8507)
36f64a9
unverified
Exclude torch==1.12.0, torchvision==0.13.0 (Fix #8395) (#8497)
1ab23fc
unverified
Junya Morioka
commited on