In the “Training output” I saw this:
/usr/local/lib/python3.8/dist-packages/sklearn/metrics/classification.py:1436: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.
precision = _prf_divide(tp_sum, pred_sum,
My ID is: 79256
This Python warning is usually due to not enough data uploaded to your project, as it happens when there is a point in your confusion matrix that divides by zero.
You will need to upload more data, or train your model further.
– Jenny
That is strange because I have uploaded more then 1000 pictures and there is more then 100 pictures for each tag.
How should I train the model to make this work?
Thanks @andreas.bomholtz for the report. I thought we had suppressed all these warnings sorry 
Do you have any more lines of calling stack trace by any chance?
Cheers, Mat
Hi Mat,
No that was all the call stack that was printet.
Best regards,
Andreas