Open
Description
Hello. I am encountering error "Training failed due to num_samples should be a positive integer value, but got num_samples=0:" when running the train.py script on my custom images. The folder with my data exists and is recognized in the script at the path I indicate in the yaml location file.
I checked online and some suggested training sample may be too small. However it is around 4,400 images in my case.
Any idea what this issue may be caused by?
Many thanks in advance.
Metadata
Metadata
Assignees
Labels
No labels