Training method enquiry and evaluation issue

[marypilataki]

Hello,

Thank you for providing the PyTorch version of onsets-and-frames. I would like to ask 2 questions.

• Is the training method here similar to the one described on the onsets and frames paper?

we split the training audio into smaller files...
We found that 20 second splits allowed us to achieve a reasonable
batch size during training of at least 8...
When notes are active and we must split, we
choose a zero-crossing of the audio signal. Inference is
performed on the original and un-split audio file.

If not, could you please explain the inputs and predictions of the model?

• When using the checkpoint provided and running the evaluation, I get zero predictions for both MAPS and MAESTRO. I have updated the mir_eval library to the latest version and still having the same issue. Could you please advise me on this?

UserWarning: Reference notes are empty.
UserWarning: Estimated notes are empty.
UserWarning: Estimate frequencies are all empty.
UserWarning: Reference frequencies are all empty.
UserWarning: Reference frequencies are all empty.

Thank you!

[healthy-pod]

I also had the same issue even after updating mir_eval. It was fixed when I increased the number of iterations from 1000 to 1000000 (takes forever without a GPU so I am using a GPU now). I don't intend to let it run to the end so I set checkpoint_interval in train.py (see below [1]) to 10000 to get a checkpoint/model that I can evaluate every 10000 iterations. Evaluating the model at the first 10000-iterations checkpoint went as expected so I can confirm at least that works.

[1] https://github.com/jongwook/onsets-and-frames/blob/master/train.py#L26