search

ZZWaang / polyphonic-chord-texture-disentanglement

The repository of the paper: Wang et al., Learning interpretable representation for controllable polyphonic music generation, ISMIR 2020.
MIT License
41 stars 12 forks source link

Fixed code with inference file, dataset, and some other essential files #4

Open CloudBao17 opened 6 months ago

CloudBao17 commented 6 months ago

polyphonic-chord-texture-disentanglement-master.zip Last few months I tried to replicate this paper and I found a few bugs inside the code. Also, it lacked some essential files like 'pop909-4-bin', 'ind.pkl', and the whole inference part. Therefore, I fixed those small bugs and wrote inference file based on https://github.com/ZZWaang/icm-deep-music-generation. Here I provide the whole work including all these files mentioned above. I hope he or she can get a better comprehension on this work and meet less problem with the code if there's some musicians with not strong code background trying to replicate this paper like me in the future.

hua2014 commented 5 months ago

Well you have done a very good job, I am doing some research in this field, here I have a question about essential files 'pop909-4-bin', can you guide me to how the files in this folder are generated?

polyphonic-chord-texture-disentanglement-master.zip Last few months I tried to replicate this paper and I found a few bugs inside the code. Also, it lacked some essential files like 'pop909-4-bin', 'ind.pkl', and the whole inference part. Therefore, I fixed those small bugs and wrote inference file based on https://github.com/ZZWaang/icm-deep-music-generation. Here I provide the whole work including all these files mentioned above. I hope he or she can get a better comprehension on this work and meet less problem with the code if there's some musicians with not strong code background trying to replicate this paper like me in the future.

CloudBao17 commented 5 months ago

Hi，hua！Sorry to say I didn’t write code to deal with the original pop909 dataset. The file ‘pop9094bin’ is obtained from one of this paper’s authors. I also have some other versions of this dataset like pop9096bin etc. However, they told me that it’s not hard to cut pop909 dataset into bars using some common tools like pretty-midi. If you want to write code to deal with the original dataset on your own, maybe you can check pretty-midi on the website or just ask for gpt-4 :). If you want some other versions of pop909, just let me know and I will send you. Hope your research goes well!

hua2014 @.***>

Well you have done a very good job, I am doing some research in this field, here I have a question about essential files 'pop909-4-bin', can you guide me to how the files in this folder are generated?

polyphonic-chord-texture-disentanglement-master.zip https://github.com/ZZWaang/polyphonic-chord-texture-disentanglement/files/14550171/polyphonic-chord-texture-disentanglement-master.zip Last few months I tried to replicate this paper and I found a few bugs inside the code. Also, it lacked some essential files like 'pop909-4-bin', 'ind.pkl', and the whole inference part. Therefore, I fixed those small bugs and wrote inference file based on https://github.com/ZZWaang/icm-deep-music-generation. Here I provide the whole work including all these files mentioned above. I hope he or she can get a better comprehension on this work and meet less problem with the code if there's some musicians with not strong code background trying to replicate this paper like me in the future.

— Reply to this email directly, view it on GitHub https://github.com/ZZWaang/polyphonic-chord-texture-disentanglement/issues/4#issuecomment-2095641525, or unsubscribe https://github.com/notifications/unsubscribe-auth/BG2XZAJ4TV3AEJL6EM4IRC3ZA5KDNAVCNFSM6AAAAABEO6QIGCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOJVGY2DCNJSGU . You are receiving this because you authored the thread.Message ID: <ZZWaang/polyphonic-chord-texture-disentanglement/issues/4/2095641525@ github.com>