fix: requirements.txt to reduce vulnerabilities
The following vulnerabilities are fixed by pinning transitive dependencies:
- https://snyk.io/vuln/SNYK-PYTHON-WHEEL-3180413
Co-authored-by: snyk-bot <snyk-bot@snyk.io>
* A minor correction in a comment
I added the 'h' in 'https' in the link to the label smoothing issue.
Signed-off-by: Kumar Selvakumaran <62794224+kumar-selvakumaran@users.noreply.github.com>
* Auto-format by https://ultralytics.com/actions
---------
Signed-off-by: Kumar Selvakumaran <62794224+kumar-selvakumaran@users.noreply.github.com>
Co-authored-by: UltralyticsAssistant <web@ultralytics.com>
* Update dataloaders.py
This is to address (and hopefully fix) this issue: Multi-GPU DDP RAM multiple-cache bug #3818 (https://github.com/ultralytics/yolov5/issues/3818). This was a very serious and "blocking" issue until I could figure out what was going on. The problem was especially bad when running Multi-GPU jobs with 8 GPUs, RAM usage was 8x higher than expected (!), causing repeated OOM failures. Hopefully this fix will help others.
DDP causes each RANK to launch it's own process (one for each GPU) with it's own trainloader, and its own RAM image cache. The DistributedSampler used by DDP (https://github.com/pytorch/pytorch/blob/master/torch/utils/data/distributed.py) will feed only a subset of images (1/WORLD_SIZE) to each available GPU on each epoch, but since the images are shuffled between epochs, each GPU process must still cache all images. So I created a subclass of DistributedSampler called SmartDistributedSampler that forces each GPU process to always sample the same subset (using modulo arithmetic with RANK and WORLD_SIZE) while still allowing random shuffling between epochs. I don't believe this disrupts the overall "randomness" of the sampling, and I haven't noticed any performance degradation.
Signed-off-by: davidsvaughn <davidsvaughn@gmail.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update dataloaders.py
move extra parameter (rank) to end so won't mess up pre-existing positional args
* Update dataloaders.py
removing extra '#'
* Update dataloaders.py
sample from DDP index array (self.idx) in mixup mosaic
* Merging self.indices and self.idx (DDP indices) into single attribute (self.indices).
Also adding SmartDistributedSampler to segmentation dataloader
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Multiply GB displayed by WORLD_SIZE
---------
Signed-off-by: davidsvaughn <davidsvaughn@gmail.com>
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>
* Added ClearML instance segmentation and classification support
* Cleaned up ClearML plot output
* typos
* Log results as plots instead of debug samples
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>
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>
* improving evole in train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix gen_ranges value in mutation part.
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* fix invalid syntax in line 532
remove on tab from "else"
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* Update train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* fix range index
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* Update train.py
fix population size
add crossover min and max rate
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update comments
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* save population for last generation
The latest version incorporates a significant update whereby all hyper parameters are now stored in the population section of "evolve_population.yaml," located in "yolov5\data\hyps," following the transition to the new generation. This development allows for the continuation of a previously abandoned evolution process by utilizing the former population. Additionally, a new argument, "--evolve_population," has been introduced to enable the relocation of the manual "evolve_population.yaml" to any project directory to load for the aforementioned purpose. This enhancement offers greater flexibility and convenience to the users, making it easier for them to resume their evolutionary process.
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove try - except
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update train.py
Add resume resume_evolve arg for **resume evolve from last generation**.
Population will load from data/hyp by default and load all yaml file form them.
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* Update train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* Update train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* Update train.py
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* Update README.zh-CN.md
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
* Update train.py
update pop_size
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
---------
Signed-off-by: Shayan Mousavinia <45814390+ShAmoNiA@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* added imagenet small versions 10,100 and 1000
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
---------
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Glenn Jocher
inisis
Nick Martin
dependabot[bot]
Abhishek Sharma
Kumar Selvakumaran
Hongbo
imyhxy
Paula Derrenger
Adrian Boguszewski
Muhammad Rizwan Munawar
Aarni Koskela
davidsvaughn
Victor Sonck
Shayan Mousavinia
Nouran Ali