* Add PyTorch AMP check
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Cleanup
* Cleanup
* Cleanup
* Robust for DDP
* Fixes
* Add amp enabled boolean to check_train_batch_size
* Simplify
* space to prefix
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Improve mAP0.5-0.95
Two changes provided
1. Added limit on the maximum number of detections for each image likewise pycocotools
2. Rework process_batch function
Changes #2 solved issue #4251
I also independently encountered the problem described in issue #4251 that the values for the same thresholds do not match when changing the limits in the torch.linspace function.
These changes solve this problem.
Currently during validation yolov5x.pt model the following results were obtained:
from yolov5 validation
Class Images Labels P R mAP@.5 mAP@.5:.95: 100%|██████████| 157/157 [01:07<00:00, 2.33it/s]
all 5000 36335 0.743 0.626 0.682 0.506
from pycocotools
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.505
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.685
These results are very close, although not completely pass the competition issue #2258.
I think it's problem with false positive bboxes matched ignored criteria, but this is not actual for custom datasets and does not require an additional solution.
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Remove line to retain pycocotools results
* Update val.py
* Update val.py
* Remove to device op
* Higher precision int conversion
* Update val.py
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Create docker.yml
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update docker.yml
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update docker.yml
* Update docker.yml
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Cleanup
* Cleanup2
* [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>
* Add `@threaded` decorator
* [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>
* Pin downloads to release version
Fixes a release version to avoid forward-compatibility issues in future releases.
* [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>
* Ability to dowlnoad older assets
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Cleanup
* Cleanup2
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>
* FROM nvcr.io/nvidia/pytorch:22.04-py3
* Update Docker
* Update Docker
* Update Docker
* Update Docker
* Update TRT auto-install
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Cleanup
* Cleanup
* Cleanup cpu
* Cleanup cpu
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* PyTorch Hub `_verbose=False` fix2
* Update downloads.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update hubconf.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update
* Update
* Update
* Update
* Update
* Update
* Update
* Update
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Update restapi.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update restapi.py
* Update restapi.py
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update restapi.py
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
* Add redundant weights download mirrors
This PR seeks to improve access worldwide to YOLOv5 weights. Universal access to AI for all is our core value, and we are against any censorship or restriction efforts.
I've uploaded the official YOLOv5 v6.1 weights to a primary backup bucket on GCP and to secondary backup on Google Drive at https://drive.google.com/drive/folders/1EFQTEUeXWSFww0luse2jB9M1QNZQGwNl. Autodownload with try the first two locations (GitHuB release assets and GCP bucket), and point users to the Google Drive folder if the first two fail.
* [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>
* support nomedia
* support nomedia for validation
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update train.py
* Revert no plot evolve
evolve plots do not contain any images
* Revert plot_results
contains no media
* Update wandb_utils.py
* sync-bn cleanup
* Cleanup
* Rename nomedia -> noplots
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>
* Update .pre-commit-config.yaml
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update .pre-commit-config.yaml
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update CONTRIBUTING.md
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update README.md
* Update README.md
* Update README.md
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 a way to skip dependency auto-installation.
Setting the environment variable `YOLOv5_AUTOINSTALL=False` will
skip installing any missing dependencies as if the user had passed
`install=False` to `check_requirements`.
* Cleanup
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
* Update `git_describe()`
Add .git path check to avoid `fatal: not a git repository (or any of the parent directories): .git` printout
* Update general.py
* add version warning
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Cleanup
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Update __init__.py
* [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>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
* Print dataset scan only `if RANK in (-1, 0)`
* [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>
* Simplify callbacks.py return
* Indent args (pytorch convention)
* [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>
* Copy wandb param dict before training to avoid overwrites.
Copy the hyperparameter dict retrieved from wandb configuration before passing it to `train()`. Training overwrites parameters in the dictionary (eg scaling obj/box/cls gains), which causes the values reported in wandb to not match the input values. This is confusing as it makes it hard to reproduce a run, and also throws off wandb's Bayesian sweep algorithm.
* Cleanup
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
