Katpeeler/midi_model_3
Text Generation
•
Updated
•
15
•
1
The dataset viewer is taking too long to fetch the data. Try to refresh this page.
Server-side error
Error code: ClientConnectionError
YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/datasets-cards)
JSFakes (Dr. Tristan Behrens).
This is a tokenized version of the JS-Fakes dataset by Omar Peracha. The original dataset can be found here: js-fakes.git
The representation is four tracks with four bars per track.
Purpose.
This dataset is a good starting point for Music Generation. You could train GPT-2 on the samples to compose music.
Contact.
Find me on LinkedIn and say hello.
If you find and issue or have a feature request, please contact me.
Please be so kind and like this dataset if you find it useful.
- Downloads last month
- 77
Size of downloaded dataset files:
13.7 MB
Size of the auto-converted Parquet files:
1.87 MB
Number of rows:
4,479
Models trained or fine-tuned on TristanBehrens/js-fakes-4bars