Deeplabcut: IndexError running "Step1_EvaluateModelonDataset.py" (Run training until at least 1 snapshot is stored before proceeding to "Step1_EvaluateModelonDataset.py")

[cathy-liu23]

The error produced is:
" File "/data/home/jli819/deeplabcut/DeepLabCut/Evaluation-Tools/Step1_EvaluateModelonDataset.py", line 68, in
cfg['init_weights'] = os.path.join(modelfolder,'train',Snapshots[snapIndex])
IndexError: index -1 is out of bounds for axis 0 with size 0"

We saw the other issue with an IndexError running the same file, but their solution did not fix our problem.

Did anyone else also run into this error? If so, what is a possible fix? Error occurred on Red Hat Enterprise Linux Server release 7.5 (Maipo).

Thanks!!

[MMathisLab]

Can you double check that you saved snapshot(s)? As default, the save_iters is set to 50,000 so if you stopped training before that point you would not have anything saved. (You should see them in the models/train folder)

[cathy-liu23]

Thank you so much for the quick reply and for solving our problem! We stopped training before the 50,000 cutoff, so we didn't have the saved snapshot(s). Thank you again!

[MMathisLab]

no problem, happy DeepLabCutting!