AssertionError: Image Not Found ..\coco128\images\train2017\000000000357.jpg

[HerrLieber]

❔Question

When I tried to train coco128 using the environment created on colab using this link:

https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb

I encountered the following problem when I followed the steps step by step. I have tried to retrain many times and each time this is the error.

AssertionError: Image Not Found ..\coco128\images\train2017\000000000357.jpg

As you can see from the image, this image exists under this path.

Additional context

In addition two bug files were generated at runtime, how do I upload them to this screen?

[HerrLieber]

❔Question

When I tried to train coco128 using the environment created on colab using this link:

https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb

I encountered the following problem when I followed the steps step by step. I have tried to retrain many times and each time this is the error.

AssertionError: Image Not Found ..\coco128\images\train2017\000000000357.jpg

As you can see from the image, this image exists under this path.

Additional context

In addition two bug files were generated at runtime, how do I upload them to this screen?

debug.log
debug-internal.log

[glenn-jocher]

@HerrLieber I am not able to reproduce your error, tested right now in Colab notebook. Steps to train are simply:

1. Run Setup cell
2. Run Train cell

[NotFound1911]

I encountered the same error on the server. So I changed the path to an absolute path.
utils/datasets.py
...
# self.img_files = sorted([x.replace('/', os.sep) for x in f if x.split('.')[-1].lower() in img_formats])
self.img_files = f
...
# self.img_files = list(cache.keys()) # update
...
Maybe you can have a try.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[yicong-q]

When we are training, a (.cache) file will be generated under the label folder.
This file is generated to facilitate the next training.
We only need to delete this (.cache) file to solve this problem.

[glenn-jocher]

Deleting the .cache file can indeed resolve path-related issues during training. If the problem persists, ensure you're using the latest YOLOv5 version.

[yicong-q]

Deleting the .cache file can indeed resolve path-related issues during training. If the problem persists, ensure you're using the latest YOLOv5 version.

The above method has solved this problem.
Thank you for still tracking this issue.

[glenn-jocher]

Glad to hear that resolved your issue! If you have any more questions, feel free to ask.