val.py --task speed parser parameter return nothing

[Haebuk]

Search before asking

• I have searched the YOLOv5 issues and found no similar bug report.

YOLOv5 Component

Validation

Bug

When excuting val.py with --task speed, the --name parser parameter is ignored and return nothing.

$python yolov5/val.py --data VisDrone.yaml --name bug-testing --weights YOLOv5/yolov5n/weights/best.pt --task speed

...

Results saved to yolov5/runs/val/exp5

Environment

• YOLO: YOLOv5 🚀 v6.0-74-gb8f979b torch 1.10.0+cu113 CUDA:0 (NVIDIA GeForce RTX 3080 Ti, 12051MiB)
• OS: Ubuntu 20.
• Python: 3.8.12

Minimal Reproducible Example

python val.py --data VisDrone.yaml --task speed --name bug-test

> Result saved to yolov5/runs/val/exp (not bug-test, and there is no file in the folder)

Additional

When excuting the shell without --task parser, It returns well.

Are you willing to submit a PR?

• Yes I'd like to help by submitting a PR!

[glenn-jocher]

@Haebuk thanks for the bug report! --project and --name should always be used to create your save_dir = project/name. I'll take a look.

[glenn-jocher]

@Haebuk good news 😃! Your original issue may now be fixed ✅ in PR #5608. This PR updated val.py to pass all arguments by default to all tasks. Note though that --task speed will not produce any saved output, the results are simply displayed to screen.

There was a separate PR #5341 which attempted to log console outputs to text file also, but this is still in progress.

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 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 🚀!