Training on midi

[ep-oi]

I want to train a S4 model on midi, which I already have as discrete events, for example in the form of .csv. How do I create a dataset with which I can train?
In other words, how can I create a dataset consisting of three columns in a .csv files or how can I train such files with python -m train?
I would be very grateful to get an answer as soon as possible.

This is what the csv data looks like:
960 | 70 | 0
0 | 70 | 74
120 | 70 | 0
120 | 70 | 70
120 | 70 | 0

[ep-oi]

I got this to work. However !python -m generate dataset=bell checkpoint_path=/content/outputs/2022-12-14/23-40-07-965990/checkpoints/loss.ckpt raises a problem:

File "/content/generate.py", line 222, in main
    batch = (x.repeat(config.n_reps, 1), None, None)
RuntimeError: Number of dimensions of repeat dims can not be smaller than number of dimensions of tensor

[albertfgu]

It looks like the data doesn't have the shape that the generate script expects. Try running one of the documented examples and setting a breakpoint to check the shapes.

[albertfgu]

By the way, I'll note that MIDI is a data format that I think S4 will have a hard time on. In general I think CNNs struggle with data that is not time-invariant, especially "event-encoded" data like MIDI.