github-actions[bot]
commented
3 years ago 👋 Hello @mattpopovich, 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.txtEnvironments
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.
mattpopovich
glenn-jocher
🐛 Bug
I have a system with 3x 1080 GPU's. When I attempt to run multi-gpu training, it seems to load the model onto the GPUs (can tell from
nvidia-smi), but hangs before actually training. I've been able to confirm that the command I'm using works for multi-gpu training on a 4x V100 system, so I believe that the command is correct. Both are running on the yolov5 docker container.To Reproduce
1) Enter docker container 2) Begin multi-gpu training on system with 3x GPUs
Output
Another thing I saw that was interesting was if I try to train on 2 GPUs but use a batch that is only divisible by 2 (and not 3), it errors out telling me that the batch is not divisible by 3 (which shouldn't matter because I'm trying to train on 2 GPUs, not 3 GPUs)
Expected behavior
I expected yolov5 to also show me a description of the model's layers being loaded onto the GPUs then to begin training.
Environment
If applicable, add screenshots to help explain your problem.
Additional context
What I see from
nvidia-smiwhen attempting to train on 2/3 GPUs with batch of 12 when yolov5 is stuck and frozen/hanging. Seems like GPU0 doesn't finish initializing? It has half the memory usage of GPU1.This is what I see on PC2 once the model has been loaded (and before it begins training)