Glenn Jocher
b2194b9015
`yolo.py --profile` updates ( #7170 )
2年前
Glenn Jocher
7830e91b9a
`yolo.py --profile` default GPU batch size 16
2年前
Glenn Jocher
3373aab56c
NMS unused variable fix ( #7161 )
* NMS unused variable fix
* Update general.py
2年前
Glenn Jocher
e19f87eb4b
Sidestep `os.path.relpath()` Windows bug ( #7158 )
* Sidestep os.path.relpath() Windows bug
os.path.relpath() seems to have a major bug on Windows due to Windows horrible path handling. This fix attempts to sidestep the issue.
```
File "C:\Users\mkokg/.cache\torch\hub\ultralytics_yolov5_master\export.py", line 64, in
ROOT = Path(os.path.relpath(ROOT, Path.cwd())) # relative
File "C:\Users\mkokg\AppData\Local\Programs\Python\Python310\lib\ntpath.py", line 718, in relpath
raise ValueError("path is on mount %r, start on mount %r" % (
ValueError: path is on mount 'C:', start on mount 'D:'
```
* Update yolo.py
* Update yolo.py
* Update yolo.py
* Update export.py
2年前
Glenn Jocher
26bfd44465
Adjust NMS time limit warning to batch size ( #7156 )
2年前
Glenn Jocher
7a2a11893b
Add Architecture Summary to README Tutorials ( #7146 )
* Add Architecture Summary to README Tutorials
Per https://github.com/ultralytics/yolov5/issues/6998#issuecomment-1073517405
* Update README.md
2年前
Zengyf-CVer
a4c661873f
Fix `detect.py --view-img` for non-ASCII paths ( #7093 )
* Update detect.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update general.py
* Update detect.py
* Update general.py
* Update general.py
* Update general.py
* Update general.py
* Update general.py
* Update general.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update general.py
* Update general.py
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2年前
RcINS
d115bbf509
Fix `cv2.imwrite` on non-ASCII paths ( #7139 )
* Fix imwrite on non-ASCII paths
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update general.py
* Update __init__.py
* Update __init__.py
* Update datasets.py
* Update hubconf.py
* Update detect.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update general.py
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2年前
Glenn Jocher
a0a4adf6de
Add PyTorch Hub `results.save(labels=False)` option ( #7129 )
Resolves https://github.com/ultralytics/yolov5/issues/388#issuecomment-1077121821
2年前
yeshanliu
bc3ed957ce
`np.fromfile()` Chinese image paths fix ( #6979 )
* :tada: :new: now can read Chinese image path.
use "cv2.imdecode(np.fromfile(f, np.uint8), cv2.IMREAD_COLOR)" instead of "cv2.imread(f)" for Chinese image path.
* Update datasets.py
* Update __init__.py
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2年前
Glenn Jocher
c3ae4e4af6
Multi-threaded VisDrone and VOC downloads ( #7108 )
* Multi-threaded VOC download
* Update VOC.yaml
* Update
* Update general.py
* Update general.py
2年前
Glenn Jocher
ecc2c7ba73
Remove named arguments where possible ( #7105 )
* Remove named arguments where possible
Speed improvements.
* Update yolo.py
* Update yolo.py
* Update yolo.py
2年前
Glenn Jocher
6134ec5d94
Model summary `pathlib` fix ( #7104 )
Stems not working correctly for YOLOv5l with current .rstrip() implementation. After fix:
```
YOLOv5l summary: 468 layers, 46563709 parameters, 46563709 gradients, 109.3 GFLOPs
```
2年前
Glenn Jocher
ee0b3b2a95
Update benchmarks significant digits ( #7103 )
2年前
Glenn Jocher
05aae17333
`torch.split()` 1.7.0 compatibility fix ( #7102 )
* Update loss.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update loss.py
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
2年前
Glenn Jocher
a600baed8e
Update model summary to display model name ( #7101 )
2年前
Glenn Jocher
a2d617ece9
Update loss for FP16 `tobj` ( #7088 )
2年前
Glenn Jocher
6f128031d0
Update loss.py with `if self.gr < 1:` ( #7087 )
* Update loss.py with `if self.gr < 1:`
* Update loss.py
2年前
Glenn Jocher
9b771a3e71
Revert "Update detect.py non-inplace with `y.tensor_split()` ( #7062 )" ( #7074 )
This reverts commit d5e363f29d
.
2年前
Glenn Jocher
9cd89b75cc
Fix2 `check_anchor_order()` in pixel-space not grid-space ( #7067 )
Follows https://github.com/ultralytics/yolov5/pull/7060 which provided only a partial solution to this issue. #7060 resolved occurences in yolo.py, this applies the same fix in autoanchor.py.
2年前
Glenn Jocher
178c109576
Add non-zero `da` `check_anchor_order()` condition ( #7066 )
2年前
Glenn Jocher
9e75cbf4c1
Update __init__.py ( #7065 )
2年前
Glenn Jocher
e278fd63ec
Update W&B message to `LOGGER.info()` ( #7064 )
2年前
Glenn Jocher
0529b77232
Update common.py lists for tuples ( #7063 )
Improved profiling.
2年前
Glenn Jocher
d5e363f29d
Update detect.py non-inplace with `y.tensor_split()` ( #7062 )
2年前
Glenn Jocher
f327eee614
Fix `check_anchor_order()` in pixel-space not grid-space ( #7060 )
* Update `check_anchor_order()`
Use mean area per output layer for added stability.
* Check in pixel-space not grid-space fix
2年前
Philip Gutjahr
529fbc1814
Use PIL to eliminate chroma subsampling in crops ( #7008 )
* use pillow to save higher-quality jpg (w/o color subsampling)
* Cleanup and doc issue
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2年前
Glenn Jocher
9f4d11379b
Fix incomplete URL substring sanitation ( #7056 )
Resolves code scanning alert in https://github.com/ultralytics/yolov5/issues/7055
2年前
Glenn Jocher
6843ea5d7f
Create SECURITY.md ( #7054 )
* Create SECURITY.md
Resolves https://github.com/ultralytics/yolov5/issues/7052
* Move into ./github
* Update SECURITY.md
2年前
Glenn Jocher
9ebec7885f
Update Dockerfile to `git clone` instead of `COPY` ( #7053 )
Resolves git command errors that currently happen in image, i.e.:
```bash
root@382ae64aeca2:/usr/src/app# git pull
Warning: Permanently added the ECDSA host key for IP address '140.82.113.3' to the list of known hosts.
git@github.com: Permission denied (publickey).
fatal: Could not read from remote repository.
Please make sure you have the correct access rights
and the repository exists.
```
2年前
Glenn Jocher
b0ba101ac0
`ComputeLoss()` indexing/speed improvements ( #7048 )
* device as class attribute
* Update loss.py
* Update loss.py
* improve zeros
* tensor split
2年前
Mrinal Jain
4effd064b1
Consistent saved_model output format ( #7032 )
2年前
Max Strobel
7c6a33564a
fix: add default PIL font as fallback ( #7010 )
* fix: add default font as fallback
Add default font as fallback if the downloading of the Arial.ttf font
fails for some reason, e.g. no access to public internet.
* Update plots.py
Co-authored-by: Maximilian Strobel <Maximilian.Strobel@infineon.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2年前
Glenn Jocher
3f634d43c8
Conditional `Timeout()` by OS (disable on Windows) ( #7013 )
* Conditional `Timeout()` by OS (disable on Windows)
* Update general.py
2年前
Glenn Jocher
c09fb2aa95
Update TQDM bar format ( #6988 )
2年前
Glenn Jocher
932dc78496
YOLOv5 Export Benchmarks for GPU ( #6963 )
* Add benchmarks.py GPU support
* Updates
* Updates
* Updates
* Updates
* Add --half
* Add TRT requirements
* Cleanup
* Add TF to warmup types
* Update export.py
* Update export.py
* Update benchmarks.py
2年前
Glenn Jocher
99de551f97
pt model to cpu on TF export
2年前
Glenn Jocher
2d45de617e
Model `ema` key backward compatibility fix ( #6972 )
Fix for older model loading issue in d3d9cbce22 (commitcomment-68622388)
2年前
paradigm
c13d4ce7ef
EdgeTPU optimizations ( #6808 )
* removed transpose op for better edgetpu support
* fix for training case
* enabled experimental new quantizer flag
* precalculate add and mul ops at compile time
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2年前
Glenn Jocher
701e1177ac
Tensor initialization on device improvements ( #6959 )
* Update common.py speed improvements
Eliminate .to() ops where possible for reduced data transfer overhead. Primarily affects warmup and PyTorch Hub inference.
* Updates
* Updates
* Update detect.py
* Update val.py
2年前
Glenn Jocher
52c1399fdc
DetectMultiBackend() return `device` update ( #6958 )
Fixes ONNX validation that returns outputs on CPU.
2年前
Glenn Jocher
c84dd27d62
New val.py `cuda` variable ( #6957 )
* New val.py `cuda` variable
Fix for ONNX GPU val.
* Update val.py
2年前
Glenn Jocher
c6b4f84fd1
Update Dockerfile `torch==1.11.0+cu113` ( #6954 )
2年前
Glenn Jocher
b94b59e199
DetectMultiBackend() `--half` handling ( #6945 )
* DetectMultiBackend() `--half` handling
* CI fixes
* rename .half to .fp16 to avoid conflict
* warmup fix
* val update
* engine update
* engine update
2年前
Glenn Jocher
84efa62b2d
Fix PyTorch Hub export inference shapes ( #6949 )
May resolve https://github.com/ultralytics/yolov5/issues/6947
2年前
Glenn Jocher
caf7ad0500
Allow 3-point segments ( #6938 )
May resolve https://github.com/ultralytics/yolov5/issues/6931
2年前
Glenn Jocher
055e72af5b
Optimize PyTorch 1.11.0 compatibility update ( #6933 )
2年前
Glenn Jocher
d3d9cbce22
PyTorch 1.11.0 compatibility updates ( #6932 )
Resolves `AttributeError: 'Upsample' object has no attribute 'recompute_scale_factor'` first raised in https://github.com/ultralytics/yolov5/issues/5499
2年前
Glenn Jocher
6dd82c0252
Move `git_describe()` to general.py ( #6918 )
* Move `git_describe()` to general.py
* Move `git_describe()` to general.py
2年前
Glenn Jocher
e6e36aac10
Update bytes to GB with bitshift ( #6886 )
2年前