Question about loading pretrained model #3494

LIU-FAYANG · 2021-06-07T07:28:33Z

❔Question

HI, thanks for your great work. I tried to load the pretrianed model with torch.hub.load() and it kept installing opencv-python, which is not desired because the object tracker I'm using only support opencv-contrib-python. Is there any method to avoid installing opencv-python when loading the pretrained models?

Also, I tied use detector = torch.load('yolov5s.pt',map_location=torch.device('cpu'))['model'] to load the model locally and some error showed up: do you have any suggestions to solve this?
Thanks for your kind help!:)

Traceback (most recent call last):
File "loitering_main.py", line 188, in main
dets_ = detector(imgs=frame[:,:,::-1]) # reverse the channels to RGB format for torch model since opencv is in BGR format
File "/home/psl/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
TypeError: forward() got an unexpected keyword argument 'imgs'

Additional context

github-actions · 2021-06-07T07:29:21Z

👋 Hello @LIU-FAYANG, thank you for your interest in 🚀 YOLOv5! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.

If this is a 🐛 Bug Report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://www.ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.7. To install run:

$ pip install -r requirements.txt

Environments

YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):

Google Colab and Kaggle notebooks with free GPU:
Google Cloud Deep Learning VM. See GCP Quickstart Guide
Amazon Deep Learning AMI. See AWS Quickstart Guide
Docker Image. See Docker Quickstart Guide

Status

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), testing (test.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

Fix for 3rd party or contrib versions of installed OpenCV as in #3494.

glenn-jocher · 2021-06-07T08:04:42Z

@LIU-FAYANG good news 😃! Your original issue may now be fixed ✅ in PR #3495. This PR adds opencv-python to the hubconf.py check_requirements() exclude list, allowing for 3rd party or contrib versions of OpenCV to be used. To receive this update:

Git – git pull from within your yolov5/ directory or git clone https://github.com/ultralytics/yolov5 again
PyTorch Hub – Force-reload with model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
Notebooks – View updated notebooks
Docker – sudo docker pull ultralytics/yolov5:latest to update your image

Thank you for spotting this issue and informing us of the problem. Please let us know if this update resolves the issue for you, and feel free to inform us of any other issues you discover or feature requests that come to mind. Happy trainings with YOLOv5 🚀!

LIU-FAYANG · 2021-06-07T09:07:52Z

Hi, I tried to git clone again in current repository and change the code to:
detector = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload = True) and the terminal still shows:

requirements: opencv-python>=4.1.2 not found and is required by YOLOV5, attempting auto-update.

glenn-jocher · 2021-06-07T09:39:09Z

@LIU-FAYANG hi, I'm sorry about that. Yes I think my update only propagated to develop branch when PyTorch Hub is pulling master branch by default. I'll push an identical update to master shortly which should fix this for you.

@SkalskiP @kalenmike this example issue.

SkalskiP · 2021-06-07T09:45:07Z

@glenn-jocher Can I get some more information on why this is happening?

glenn-jocher · 2021-06-07T10:08:59Z

@SkalskiP Under the hood the torch.hub.load() command is cloning the repo that's passed in the first argument, i.e. 'ultralytics/yolov5', searching for hubconf.py at the top level (which is why we can't place it in the yolov5/ dir in the pip package), and then inside hubconf.py finding a function called 'yolov5s()', which it expects to return a model from.

model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)

The problem seems to be that torch hub is pulling master branch by default. So my fix in #3495 doesn't make it into these hub models since I merged it into develop branch.

glenn-jocher · 2021-06-07T10:10:51Z

@SkalskiP the documentation explains a bit more here:
https://pytorch.org/docs/stable/hub.html#loading-models-from-hub

SkalskiP · 2021-06-07T10:48:53Z

@glenn-jocher and why it is wrong that the torchhub is integrated with the master branch? master is a stable branch.

glenn-jocher · 2021-06-07T10:55:46Z

@SkalskiP I've fixed this problem in #3495, pushed to develop.

@LIU-FAYANG can't receive this fix because PyTorch Hub uses master.

glenn-jocher · 2021-06-07T10:56:30Z

@SkalskiP PyTorch Hub is not in the wrong, we are with our branch structure.

SkalskiP · 2021-06-07T11:04:18Z

@glenn-jocher so you merged hot-fix to develop not to master? To me "branch structure" is correct. It is simply that we should merge fixes to master and features to develop.

glenn-jocher · 2021-06-07T11:14:42Z

@SkalskiP if we merge fixes to master then everyone who git clones our repo, i.e. 2000 people a day, will not have access to the latest fixes, and will continue to raise bug reports since you've placed develop as our default branch.

SkalskiP · 2021-06-07T11:22:42Z

@glenn-jocher a, now I understand. So I did set develop as the default branch only to make sure that it will be set as PR target branch automatically. Setting develop as default is not mandatory in any way. It was only for convenience. If we feel it causes more problems let's set master as default, but then we need to remember to double-check the PR target branch. But I guess in that case that makes much more sense. So my proposal is:

Keep master <- develop <- feature_branch flow
Set master as the default branch so that it would load after git clone
Remember to set correct target branch when someone create PR
Merge fixes into master and than rebase develop on fresh master
Merge new features into develop

@glenn-jocher what do you think about that?

glenn-jocher · 2021-06-07T12:06:15Z

@SkalskiP yeah maybe that's a better route, though there might be confusion indicating to the PR submitters which branch they should rebase against.

Let's discuss all the options at our meeting tomorrow with Kalen, we'll try to figure out a solution there. Thanks!

Duplicate of #3495 merged to `develop`. This PR will be merged to `master`. Fixes #3494.

glenn-jocher · 2021-06-07T13:40:28Z

@LIU-FAYANG good news 😃! Your original issue may now be fixed ✅ in PR #3507 which I just merged to master. This PR is a duplicate PR of #3495 which was merged to develop previously, which PyTorch Hub does not use, and which was the reason my first fix attempt failed to resolve your issue. To receive this update:

Git – git pull from within your yolov5/ directory or git clone https://github.com/ultralytics/yolov5 again
PyTorch Hub – Force-reload with model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
Notebooks – View updated notebooks
Docker – sudo docker pull ultralytics/yolov5:latest to update your image

Thank you for spotting this issue and informing us of the problem. Please let us know if this update resolves the issue for you, and feel free to inform us of any other issues you discover or feature requests that come to mind. Happy trainings with YOLOv5 🚀!

@kalenmike

* update ci-testing.yml (#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * update ci-testing.yml (#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * Enable direct `--weights URL` definition (#3373) * Enable direct `--weights URL` definition @kalenmike this PR will enable direct --weights URL definition. Example use case: ``` python train.py --weights https://storage.googleapis.com/bucket/dir/model.pt ``` * cleanup * bug fixes * weights = attempt_download(weights) * Update experimental.py * Update hubconf.py * return bug fix * comment mirror * min_bytes * Update tutorial.ipynb (#3368) add Open in Kaggle badge * `cv2.imread(img, -1)` for IMREAD_UNCHANGED (#3379) * Update datasets.py * comment Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * COCO evolution fix (#3388) * COCO evolution fix * cleanup * update print * print fix * Create `is_pip()` function (#3391) Returns `True` if file is part of pip package. Useful for contextual behavior modification. ```python def is_pip(): # Is file in a pip package? return 'site-packages' in Path(__file__).absolute().parts ``` * Revert "`cv2.imread(img, -1)` for IMREAD_UNCHANGED (#3379)" (#3395) This reverts commit 21a9607. * Update FLOPs description (#3422) * Update README.md * Changing FLOPS to FLOPs. Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> * Parse URL authentication (#3424) * Parse URL authentication * urllib.parse.unquote() * improved error handling * improved error handling * remove %3F * update check_file() * Add FLOPs title to table (#3453) * Suppress jit trace warning + graph once (#3454) * Suppress jit trace warning + graph once Suppress harmless jit trace warning on TensorBoard add_graph call. Also fix multiple add_graph() calls bug, now only on batch 0. * Update train.py * Update MixUp augmentation `alpha=beta=32.0` (#3455) Per VOC empirical results #3380 (comment) by @developer0hye * Add `timeout()` class (#3460) * Add `timeout()` class * rearrange order * Faster HSV augmentation (#3462) remove datatype conversion process that can be skipped * Add `check_git_status()` 5 second timeout (#3464) * Add check_git_status() 5 second timeout This should prevent the SSH Git bug that we were discussing @kalenmike * cleanup * replace timeout with check_output built-in timeout * Improved `check_requirements()` offline-handling (#3466) Improve robustness of `check_requirements()` function to offline environments (do not attempt pip installs when offline). * Add `output_names` argument for ONNX export with dynamic axes (#3456) * Add output names & dynamic axes for onnx export Add output_names and dynamic_axes names for all outputs in torch.onnx.export. The first four outputs of the model will have names output0, output1, output2, output3 * use first output only + cleanup Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Revert FP16 `test.py` and `detect.py` inference to FP32 default (#3423) * fixed inference bug ,while use half precision * replace --use-half with --half * replace space and PEP8 in detect.py * PEP8 detect.py * update --half help comment * Update test.py * revert space Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Add additional links/resources to stale.yml message (#3467) * Update stale.yml * cleanup * Update stale.yml * reformat * Update stale.yml HUB URL (#3468) * Stale `github.actor` bug fix (#3483) * Explicit `model.eval()` call `if opt.train=False` (#3475) * call model.eval() when opt.train is False call model.eval() when opt.train is False * single-line if statement * cleanup Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * check_requirements() exclude `opencv-python` (#3495) Fix for 3rd party or contrib versions of installed OpenCV as in #3494. * Earlier `assert` for cpu and half option (#3508) * early assert for cpu and half option early assert for cpu and half option * Modified comment Modified comment * Update tutorial.ipynb (#3510) * Reduce test.py results spacing (#3511) * Update README.md (#3512) * Update README.md Minor modifications * 850 width * Update greetings.yml revert greeting change as PRs will now merge to master. Co-authored-by: Piotr Skalski <SkalskiP@users.noreply.github.com> Co-authored-by: SkalskiP <piotr.skalski92@gmail.com> Co-authored-by: Peretz Cohen <pizzaz93@users.noreply.github.com> Co-authored-by: tudoulei <34886368+tudoulei@users.noreply.github.com> Co-authored-by: chocosaj <chocosaj@users.noreply.github.com> Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> Co-authored-by: Yonghye Kwon <developer.0hye@gmail.com> Co-authored-by: Sam_S <SamSamhuns@users.noreply.github.com> Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: edificewang <609552430@qq.com>

Duplicate of ultralytics#3495 merged to `develop`. This PR will be merged to `master`. Fixes ultralytics#3494. (cherry picked from commit 3f03acb)

@kalenmike

* update ci-testing.yml (ultralytics#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * update ci-testing.yml (ultralytics#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * Enable direct `--weights URL` definition (ultralytics#3373) * Enable direct `--weights URL` definition @kalenmike this PR will enable direct --weights URL definition. Example use case: ``` python train.py --weights https://storage.googleapis.com/bucket/dir/model.pt ``` * cleanup * bug fixes * weights = attempt_download(weights) * Update experimental.py * Update hubconf.py * return bug fix * comment mirror * min_bytes * Update tutorial.ipynb (ultralytics#3368) add Open in Kaggle badge * `cv2.imread(img, -1)` for IMREAD_UNCHANGED (ultralytics#3379) * Update datasets.py * comment Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * COCO evolution fix (ultralytics#3388) * COCO evolution fix * cleanup * update print * print fix * Create `is_pip()` function (ultralytics#3391) Returns `True` if file is part of pip package. Useful for contextual behavior modification. ```python def is_pip(): # Is file in a pip package? return 'site-packages' in Path(__file__).absolute().parts ``` * Revert "`cv2.imread(img, -1)` for IMREAD_UNCHANGED (ultralytics#3379)" (ultralytics#3395) This reverts commit 21a9607. * Update FLOPs description (ultralytics#3422) * Update README.md * Changing FLOPS to FLOPs. Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> * Parse URL authentication (ultralytics#3424) * Parse URL authentication * urllib.parse.unquote() * improved error handling * improved error handling * remove %3F * update check_file() * Add FLOPs title to table (ultralytics#3453) * Suppress jit trace warning + graph once (ultralytics#3454) * Suppress jit trace warning + graph once Suppress harmless jit trace warning on TensorBoard add_graph call. Also fix multiple add_graph() calls bug, now only on batch 0. * Update train.py * Update MixUp augmentation `alpha=beta=32.0` (ultralytics#3455) Per VOC empirical results ultralytics#3380 (comment) by @developer0hye * Add `timeout()` class (ultralytics#3460) * Add `timeout()` class * rearrange order * Faster HSV augmentation (ultralytics#3462) remove datatype conversion process that can be skipped * Add `check_git_status()` 5 second timeout (ultralytics#3464) * Add check_git_status() 5 second timeout This should prevent the SSH Git bug that we were discussing @kalenmike * cleanup * replace timeout with check_output built-in timeout * Improved `check_requirements()` offline-handling (ultralytics#3466) Improve robustness of `check_requirements()` function to offline environments (do not attempt pip installs when offline). * Add `output_names` argument for ONNX export with dynamic axes (ultralytics#3456) * Add output names & dynamic axes for onnx export Add output_names and dynamic_axes names for all outputs in torch.onnx.export. The first four outputs of the model will have names output0, output1, output2, output3 * use first output only + cleanup Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Revert FP16 `test.py` and `detect.py` inference to FP32 default (ultralytics#3423) * fixed inference bug ,while use half precision * replace --use-half with --half * replace space and PEP8 in detect.py * PEP8 detect.py * update --half help comment * Update test.py * revert space Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Add additional links/resources to stale.yml message (ultralytics#3467) * Update stale.yml * cleanup * Update stale.yml * reformat * Update stale.yml HUB URL (ultralytics#3468) * Stale `github.actor` bug fix (ultralytics#3483) * Explicit `model.eval()` call `if opt.train=False` (ultralytics#3475) * call model.eval() when opt.train is False call model.eval() when opt.train is False * single-line if statement * cleanup Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * check_requirements() exclude `opencv-python` (ultralytics#3495) Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics#3494. * Earlier `assert` for cpu and half option (ultralytics#3508) * early assert for cpu and half option early assert for cpu and half option * Modified comment Modified comment * Update tutorial.ipynb (ultralytics#3510) * Reduce test.py results spacing (ultralytics#3511) * Update README.md (ultralytics#3512) * Update README.md Minor modifications * 850 width * Update greetings.yml revert greeting change as PRs will now merge to master. Co-authored-by: Piotr Skalski <SkalskiP@users.noreply.github.com> Co-authored-by: SkalskiP <piotr.skalski92@gmail.com> Co-authored-by: Peretz Cohen <pizzaz93@users.noreply.github.com> Co-authored-by: tudoulei <34886368+tudoulei@users.noreply.github.com> Co-authored-by: chocosaj <chocosaj@users.noreply.github.com> Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> Co-authored-by: Yonghye Kwon <developer.0hye@gmail.com> Co-authored-by: Sam_S <SamSamhuns@users.noreply.github.com> Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: edificewang <609552430@qq.com> (cherry picked from commit f3c3d2c)

Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics#3494.

Duplicate of ultralytics#3495 merged to `develop`. This PR will be merged to `master`. Fixes ultralytics#3494.

@kalenmike

* update ci-testing.yml (ultralytics#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * update ci-testing.yml (ultralytics#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * Enable direct `--weights URL` definition (ultralytics#3373) * Enable direct `--weights URL` definition @kalenmike this PR will enable direct --weights URL definition. Example use case: ``` python train.py --weights https://storage.googleapis.com/bucket/dir/model.pt ``` * cleanup * bug fixes * weights = attempt_download(weights) * Update experimental.py * Update hubconf.py * return bug fix * comment mirror * min_bytes * Update tutorial.ipynb (ultralytics#3368) add Open in Kaggle badge * `cv2.imread(img, -1)` for IMREAD_UNCHANGED (ultralytics#3379) * Update datasets.py * comment Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * COCO evolution fix (ultralytics#3388) * COCO evolution fix * cleanup * update print * print fix * Create `is_pip()` function (ultralytics#3391) Returns `True` if file is part of pip package. Useful for contextual behavior modification. ```python def is_pip(): # Is file in a pip package? return 'site-packages' in Path(__file__).absolute().parts ``` * Revert "`cv2.imread(img, -1)` for IMREAD_UNCHANGED (ultralytics#3379)" (ultralytics#3395) This reverts commit 67db4b6. * Update FLOPs description (ultralytics#3422) * Update README.md * Changing FLOPS to FLOPs. Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> * Parse URL authentication (ultralytics#3424) * Parse URL authentication * urllib.parse.unquote() * improved error handling * improved error handling * remove %3F * update check_file() * Add FLOPs title to table (ultralytics#3453) * Suppress jit trace warning + graph once (ultralytics#3454) * Suppress jit trace warning + graph once Suppress harmless jit trace warning on TensorBoard add_graph call. Also fix multiple add_graph() calls bug, now only on batch 0. * Update train.py * Update MixUp augmentation `alpha=beta=32.0` (ultralytics#3455) Per VOC empirical results ultralytics#3380 (comment) by @developer0hye * Add `timeout()` class (ultralytics#3460) * Add `timeout()` class * rearrange order * Faster HSV augmentation (ultralytics#3462) remove datatype conversion process that can be skipped * Add `check_git_status()` 5 second timeout (ultralytics#3464) * Add check_git_status() 5 second timeout This should prevent the SSH Git bug that we were discussing @kalenmike * cleanup * replace timeout with check_output built-in timeout * Improved `check_requirements()` offline-handling (ultralytics#3466) Improve robustness of `check_requirements()` function to offline environments (do not attempt pip installs when offline). * Add `output_names` argument for ONNX export with dynamic axes (ultralytics#3456) * Add output names & dynamic axes for onnx export Add output_names and dynamic_axes names for all outputs in torch.onnx.export. The first four outputs of the model will have names output0, output1, output2, output3 * use first output only + cleanup Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Revert FP16 `test.py` and `detect.py` inference to FP32 default (ultralytics#3423) * fixed inference bug ,while use half precision * replace --use-half with --half * replace space and PEP8 in detect.py * PEP8 detect.py * update --half help comment * Update test.py * revert space Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Add additional links/resources to stale.yml message (ultralytics#3467) * Update stale.yml * cleanup * Update stale.yml * reformat * Update stale.yml HUB URL (ultralytics#3468) * Stale `github.actor` bug fix (ultralytics#3483) * Explicit `model.eval()` call `if opt.train=False` (ultralytics#3475) * call model.eval() when opt.train is False call model.eval() when opt.train is False * single-line if statement * cleanup Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * check_requirements() exclude `opencv-python` (ultralytics#3495) Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics#3494. * Earlier `assert` for cpu and half option (ultralytics#3508) * early assert for cpu and half option early assert for cpu and half option * Modified comment Modified comment * Update tutorial.ipynb (ultralytics#3510) * Reduce test.py results spacing (ultralytics#3511) * Update README.md (ultralytics#3512) * Update README.md Minor modifications * 850 width * Update greetings.yml revert greeting change as PRs will now merge to master. Co-authored-by: Piotr Skalski <SkalskiP@users.noreply.github.com> Co-authored-by: SkalskiP <piotr.skalski92@gmail.com> Co-authored-by: Peretz Cohen <pizzaz93@users.noreply.github.com> Co-authored-by: tudoulei <34886368+tudoulei@users.noreply.github.com> Co-authored-by: chocosaj <chocosaj@users.noreply.github.com> Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> Co-authored-by: Yonghye Kwon <developer.0hye@gmail.com> Co-authored-by: Sam_S <SamSamhuns@users.noreply.github.com> Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: edificewang <609552430@qq.com>

Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics/yolov5#3494.

Duplicate of #3495 merged to `develop`. This PR will be merged to `master`. Fixes ultralytics/yolov5#3494.

@kalenmike

* update ci-testing.yml (#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * update ci-testing.yml (#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * Enable direct `--weights URL` definition (#3373) * Enable direct `--weights URL` definition @kalenmike this PR will enable direct --weights URL definition. Example use case: ``` python train.py --weights https://storage.googleapis.com/bucket/dir/model.pt ``` * cleanup * bug fixes * weights = attempt_download(weights) * Update experimental.py * Update hubconf.py * return bug fix * comment mirror * min_bytes * Update tutorial.ipynb (#3368) add Open in Kaggle badge * `cv2.imread(img, -1)` for IMREAD_UNCHANGED (#3379) * Update datasets.py * comment Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * COCO evolution fix (#3388) * COCO evolution fix * cleanup * update print * print fix * Create `is_pip()` function (#3391) Returns `True` if file is part of pip package. Useful for contextual behavior modification. ```python def is_pip(): # Is file in a pip package? return 'site-packages' in Path(__file__).absolute().parts ``` * Revert "`cv2.imread(img, -1)` for IMREAD_UNCHANGED (#3379)" (#3395) This reverts commit 21a9607e00f1365b21d8c4bd81bdbf5fc0efea24. * Update FLOPs description (#3422) * Update README.md * Changing FLOPS to FLOPs. Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> * Parse URL authentication (#3424) * Parse URL authentication * urllib.parse.unquote() * improved error handling * improved error handling * remove %3F * update check_file() * Add FLOPs title to table (#3453) * Suppress jit trace warning + graph once (#3454) * Suppress jit trace warning + graph once Suppress harmless jit trace warning on TensorBoard add_graph call. Also fix multiple add_graph() calls bug, now only on batch 0. * Update train.py * Update MixUp augmentation `alpha=beta=32.0` (#3455) Per VOC empirical results ultralytics/yolov5#3380 (comment) by @developer0hye * Add `timeout()` class (#3460) * Add `timeout()` class * rearrange order * Faster HSV augmentation (#3462) remove datatype conversion process that can be skipped * Add `check_git_status()` 5 second timeout (#3464) * Add check_git_status() 5 second timeout This should prevent the SSH Git bug that we were discussing @kalenmike * cleanup * replace timeout with check_output built-in timeout * Improved `check_requirements()` offline-handling (#3466) Improve robustness of `check_requirements()` function to offline environments (do not attempt pip installs when offline). * Add `output_names` argument for ONNX export with dynamic axes (#3456) * Add output names & dynamic axes for onnx export Add output_names and dynamic_axes names for all outputs in torch.onnx.export. The first four outputs of the model will have names output0, output1, output2, output3 * use first output only + cleanup Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Revert FP16 `test.py` and `detect.py` inference to FP32 default (#3423) * fixed inference bug ,while use half precision * replace --use-half with --half * replace space and PEP8 in detect.py * PEP8 detect.py * update --half help comment * Update test.py * revert space Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Add additional links/resources to stale.yml message (#3467) * Update stale.yml * cleanup * Update stale.yml * reformat * Update stale.yml HUB URL (#3468) * Stale `github.actor` bug fix (#3483) * Explicit `model.eval()` call `if opt.train=False` (#3475) * call model.eval() when opt.train is False call model.eval() when opt.train is False * single-line if statement * cleanup Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * check_requirements() exclude `opencv-python` (#3495) Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics/yolov5#3494. * Earlier `assert` for cpu and half option (#3508) * early assert for cpu and half option early assert for cpu and half option * Modified comment Modified comment * Update tutorial.ipynb (#3510) * Reduce test.py results spacing (#3511) * Update README.md (#3512) * Update README.md Minor modifications * 850 width * Update greetings.yml revert greeting change as PRs will now merge to master. Co-authored-by: Piotr Skalski <SkalskiP@users.noreply.github.com> Co-authored-by: SkalskiP <piotr.skalski92@gmail.com> Co-authored-by: Peretz Cohen <pizzaz93@users.noreply.github.com> Co-authored-by: tudoulei <34886368+tudoulei@users.noreply.github.com> Co-authored-by: chocosaj <chocosaj@users.noreply.github.com> Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> Co-authored-by: Yonghye Kwon <developer.0hye@gmail.com> Co-authored-by: Sam_S <SamSamhuns@users.noreply.github.com> Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: edificewang <609552430@qq.com>

Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics/yolov5#3494.

Duplicate of #3495 merged to `develop`. This PR will be merged to `master`. Fixes ultralytics/yolov5#3494.

@kalenmike

* update ci-testing.yml (#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * update ci-testing.yml (#3322) * update ci-testing.yml * update greetings.yml * bring back os matrix * Enable direct `--weights URL` definition (#3373) * Enable direct `--weights URL` definition @kalenmike this PR will enable direct --weights URL definition. Example use case: ``` python train.py --weights https://storage.googleapis.com/bucket/dir/model.pt ``` * cleanup * bug fixes * weights = attempt_download(weights) * Update experimental.py * Update hubconf.py * return bug fix * comment mirror * min_bytes * Update tutorial.ipynb (#3368) add Open in Kaggle badge * `cv2.imread(img, -1)` for IMREAD_UNCHANGED (#3379) * Update datasets.py * comment Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * COCO evolution fix (#3388) * COCO evolution fix * cleanup * update print * print fix * Create `is_pip()` function (#3391) Returns `True` if file is part of pip package. Useful for contextual behavior modification. ```python def is_pip(): # Is file in a pip package? return 'site-packages' in Path(__file__).absolute().parts ``` * Revert "`cv2.imread(img, -1)` for IMREAD_UNCHANGED (#3379)" (#3395) This reverts commit 21a9607e00f1365b21d8c4bd81bdbf5fc0efea24. * Update FLOPs description (#3422) * Update README.md * Changing FLOPS to FLOPs. Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> * Parse URL authentication (#3424) * Parse URL authentication * urllib.parse.unquote() * improved error handling * improved error handling * remove %3F * update check_file() * Add FLOPs title to table (#3453) * Suppress jit trace warning + graph once (#3454) * Suppress jit trace warning + graph once Suppress harmless jit trace warning on TensorBoard add_graph call. Also fix multiple add_graph() calls bug, now only on batch 0. * Update train.py * Update MixUp augmentation `alpha=beta=32.0` (#3455) Per VOC empirical results ultralytics/yolov5#3380 (comment) by @developer0hye * Add `timeout()` class (#3460) * Add `timeout()` class * rearrange order * Faster HSV augmentation (#3462) remove datatype conversion process that can be skipped * Add `check_git_status()` 5 second timeout (#3464) * Add check_git_status() 5 second timeout This should prevent the SSH Git bug that we were discussing @kalenmike * cleanup * replace timeout with check_output built-in timeout * Improved `check_requirements()` offline-handling (#3466) Improve robustness of `check_requirements()` function to offline environments (do not attempt pip installs when offline). * Add `output_names` argument for ONNX export with dynamic axes (#3456) * Add output names & dynamic axes for onnx export Add output_names and dynamic_axes names for all outputs in torch.onnx.export. The first four outputs of the model will have names output0, output1, output2, output3 * use first output only + cleanup Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Revert FP16 `test.py` and `detect.py` inference to FP32 default (#3423) * fixed inference bug ,while use half precision * replace --use-half with --half * replace space and PEP8 in detect.py * PEP8 detect.py * update --half help comment * Update test.py * revert space Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * Add additional links/resources to stale.yml message (#3467) * Update stale.yml * cleanup * Update stale.yml * reformat * Update stale.yml HUB URL (#3468) * Stale `github.actor` bug fix (#3483) * Explicit `model.eval()` call `if opt.train=False` (#3475) * call model.eval() when opt.train is False call model.eval() when opt.train is False * single-line if statement * cleanup Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com> * check_requirements() exclude `opencv-python` (#3495) Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics/yolov5#3494. * Earlier `assert` for cpu and half option (#3508) * early assert for cpu and half option early assert for cpu and half option * Modified comment Modified comment * Update tutorial.ipynb (#3510) * Reduce test.py results spacing (#3511) * Update README.md (#3512) * Update README.md Minor modifications * 850 width * Update greetings.yml revert greeting change as PRs will now merge to master. Co-authored-by: Piotr Skalski <SkalskiP@users.noreply.github.com> Co-authored-by: SkalskiP <piotr.skalski92@gmail.com> Co-authored-by: Peretz Cohen <pizzaz93@users.noreply.github.com> Co-authored-by: tudoulei <34886368+tudoulei@users.noreply.github.com> Co-authored-by: chocosaj <chocosaj@users.noreply.github.com> Co-authored-by: BuildTools <unconfigured@null.spigotmc.org> Co-authored-by: Yonghye Kwon <developer.0hye@gmail.com> Co-authored-by: Sam_S <SamSamhuns@users.noreply.github.com> Co-authored-by: Samridha Shrestha <samridha.shrestha@g42.ai> Co-authored-by: edificewang <609552430@qq.com>

LIU-FAYANG added the question Further information is requested label Jun 7, 2021

glenn-jocher added a commit that referenced this issue Jun 7, 2021

check_requirements() exclude opencv-python

600e0c6

Fix for 3rd party or contrib versions of installed OpenCV as in #3494.

glenn-jocher mentioned this issue Jun 7, 2021

check_requirements() exclude opencv-python #3495

Merged

glenn-jocher linked a pull request Jun 7, 2021 that will close this issue

check_requirements() exclude opencv-python #3495

Merged

glenn-jocher closed this as completed in #3495 Jun 7, 2021

glenn-jocher added a commit that referenced this issue Jun 7, 2021

check_requirements() exclude opencv-python (#3495)

90b7895

Fix for 3rd party or contrib versions of installed OpenCV as in #3494.

This comment has been minimized.

Sign in to view

glenn-jocher mentioned this issue Jun 7, 2021

check_requirements() exclude opencv-python #3506

Closed

glenn-jocher added a commit that referenced this issue Jun 7, 2021

check_requirements() exclude opencv-python

f9d19e1

Duplicate of #3495 merged to `develop`. This PR will be merged to `master`. Fixes #3494.

glenn-jocher mentioned this issue Jun 7, 2021

check_requirements() exclude opencv-python #3507

Merged

glenn-jocher linked a pull request Jun 7, 2021 that will close this issue

check_requirements() exclude opencv-python #3507

Merged

glenn-jocher added a commit that referenced this issue Jun 7, 2021

check_requirements() exclude opencv-python (#3507)

3f03acb

Duplicate of #3495 merged to `develop`. This PR will be merged to `master`. Fixes #3494.

BjarneKuehl pushed a commit to fhkiel-mlaip/yolov5 that referenced this issue Aug 26, 2022

check_requirements() exclude opencv-python (ultralytics#3495)

5f8250d

Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics#3494.

BjarneKuehl pushed a commit to fhkiel-mlaip/yolov5 that referenced this issue Aug 26, 2022

check_requirements() exclude opencv-python (ultralytics#3507)

893d371

Duplicate of ultralytics#3495 merged to `develop`. This PR will be merged to `master`. Fixes ultralytics#3494.

SecretStar112 added a commit to SecretStar112/yolov5 that referenced this issue May 24, 2023

check_requirements() exclude opencv-python (#3495)

151c8f2

Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics/yolov5#3494.

SecretStar112 added a commit to SecretStar112/yolov5 that referenced this issue May 24, 2023

check_requirements() exclude opencv-python (#3507)

bb51762

Duplicate of #3495 merged to `develop`. This PR will be merged to `master`. Fixes ultralytics/yolov5#3494.

MichaelAnderson-AI added a commit to MichaelAnderson-AI/upwork-yolo-swift that referenced this issue Oct 23, 2024

check_requirements() exclude opencv-python (#3495)

81c5a65

Fix for 3rd party or contrib versions of installed OpenCV as in ultralytics/yolov5#3494.

MichaelAnderson-AI added a commit to MichaelAnderson-AI/upwork-yolo-swift that referenced this issue Oct 23, 2024

check_requirements() exclude opencv-python (#3507)

2ba7925

Duplicate of #3495 merged to `develop`. This PR will be merged to `master`. Fixes ultralytics/yolov5#3494.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Question about loading pretrained model #3494

Question about loading pretrained model #3494

LIU-FAYANG commented Jun 7, 2021

github-actions bot commented Jun 7, 2021 •

edited by UltralyticsAssistant

Loading

glenn-jocher commented Jun 7, 2021 •

edited by UltralyticsAssistant

Loading

LIU-FAYANG commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021 •

edited

Loading

SkalskiP commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021 •

edited

Loading

glenn-jocher commented Jun 7, 2021

This comment has been minimized.

SkalskiP commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021

SkalskiP commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021

SkalskiP commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021 •

edited by UltralyticsAssistant

Loading

Question about loading pretrained model #3494

Question about loading pretrained model #3494

Comments

LIU-FAYANG commented Jun 7, 2021

❔Question

Additional context

github-actions bot commented Jun 7, 2021 • edited by UltralyticsAssistant Loading

Requirements

Environments

Status

glenn-jocher commented Jun 7, 2021 • edited by UltralyticsAssistant Loading

LIU-FAYANG commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021 • edited Loading

SkalskiP commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021 • edited Loading

glenn-jocher commented Jun 7, 2021

This comment has been minimized.

SkalskiP commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021

SkalskiP commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021

SkalskiP commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021

glenn-jocher commented Jun 7, 2021 • edited by UltralyticsAssistant Loading

github-actions bot commented Jun 7, 2021 •

edited by UltralyticsAssistant

Loading

glenn-jocher commented Jun 7, 2021 •

edited by UltralyticsAssistant

Loading

glenn-jocher commented Jun 7, 2021 •

edited

Loading

glenn-jocher commented Jun 7, 2021 •

edited

Loading

glenn-jocher commented Jun 7, 2021 •

edited by UltralyticsAssistant

Loading