However, when I directly use the json file provided with the MSD dataset I have a problem reading the format mismatch, I would like to know what kind of way the authors use to process the dataset collectively, can you provide me with the data or the script used to process the data?
Hi. We support input data in a jsonl file, you can check the Inference Guidance in README and refer to demo/inference_demo/demo.jsonl to prepare your data with required format.
However, when I directly use the json file provided with the MSD dataset I have a problem reading the format mismatch, I would like to know what kind of way the authors use to process the dataset collectively, can you provide me with the data or the script used to process the data?