README.md

---
license: cc-by-nc-sa-4.0
dataset_info:
  features:
  - name: midi_hash
    dtype: string
  - name: core_chords
    sequence: int64
  - name: core_score
    sequence: int64
  splits:
  - name: train
    num_bytes: 4385400964
    num_examples: 386363
  download_size: 586309989
  dataset_size: 4385400964
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
language:
- en
tags:
- MIDI
- cores
- MIDI cores
pretty_name: midicores
size_categories:
- 100K<n<1M
---

# MIDI cores

## 386k+ select MIDI cores from [Monster MIDI Dataset](https://huggingface.co/datasets/projectlosangeles/Monster-MIDI-Dataset)

![MIDI-Cores-Logo.jpg](https://cdn-uploads.huggingface.co/production/uploads/5f57ea2d3f32f12a3c0692e6/LfaqbIW9B8m50HXst9dcx.jpeg)

***

## Load dataset

```python
#===================================================================

from datasets import load_dataset

#===================================================================

midi_cores = load_dataset('asigalov61/Monster-Cores')
```

***

### Project Los Angeles
### Tegridy Code 2025