Open liuhuanxg opened 11 months ago
So the code requires images and txt files. The dataset initially comes in the format of .mscz files. Once we have the .mscz files, we can convert them to images and txt files. The CSV file contains URLs to each individual .mscz file that you can download. The idea is to use the CSV file with a script to download all of the .mscz files using a tool such as wget
For the full details on how to do this, please check the README!
When I open this link, it prompts me to access these three links to download resources: https://musescore-dataset.xmader.com/user.jsonl https://musescore-dataset.xmader.com/score.jsonl https://musescore-dataset.xmader.com/mscz-files.csv These files appear to be in JSONL and CSV formats, but what the code requires are images and txt files. I don't know if my understanding is incorrect? If you have a more detailed description, please tell me. thanks