adrienchaton
adrienchaton
I work on a raw audio waveform model, mainly using 1d (temporal) convolutions. After the last convolution of the decoder, I add up a single channel convolution with a large...
Thanks Chris for your answer and confirming this ! It seems also beneficial for my application Good continuation, it was a really good article to my opinion
same here bro! thank you, it happened to me before and I thought I did a mistake not saving properly now I was 100% careful and sure I had saved,...
@leonwyang @philippbayer I am having the same issues as you ... here are the current status that I also face with my own data and finetuning codes (running on DGX...
thanks for your reply and the good job done! that will be a great addition to this tool
one thought I had, could anything related to the glycine linker trick as in e.g. AF multimer, be applicable to FoldSeek?
Thank you very much. Checkpoint is helpful as I want to compare retraining from scratch or fine-tuning from the maestro checkpoint ! And indeed, your midi code seems to follow...
I am running the codes with torch==1.3.0 There would be some (minor) things to update for running on higher versions.
You'd have to update error by error, but it should not be a lot since Pytorch has not changed much in the basic functions that are used here. One more...
no, I did not use either MAESTRO or MAPS I wanted to fine-tune on a custom dataset so I made a class similar to MAPS(PianoRollAudioDataset) which reads flac audio and...