# MindBigData "MNIST of Brain Digits", Emotiv EPOC EEG: Thought-Digit Recordings (Vivancos, 2022 Train/Test Release)

## Facts

- Provider: David Vivancos (MindBigData)
- License: ODbL-1.0 (https://mindbigdata.com/opendb/index.html)
- Size: 4 files
- Format: CSV (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/DavidVivancos/MindBigData2022_MNIST_EP
- Access: public, readable without an account
- Category: Neuroscience
- Source: https://arxiv.org/abs/2212.14746
- Page: https://scigantic.com/archive/6e56b957-a7a9-536f-b4d0-7b0571b5703b
- Keywords: Neuroscience, EEG, CSV, Hugging Face

## About

Emotiv EPOC subset of the MindBigData "MNIST of Brain Digits" EEG database. Raw electroencephalography (EEG) from the MindBigData "MNIST" of Brain Digits open database, recorded with a 14-channel Emotiv EPOC consumer headset from a single subject, David Vivancos, between 2014 and 2015 while he saw a digit from 0 to 9 and thought about it, for 2 seconds per event. This is the Emotiv EPOC subset: AF3, F7, F3, FC5, T7, P7, O1, O2, P8, T8, FC6, F4, F8 and AF4 (14 channels, 10/20 locations), 256 samples per channel per 2-second event (about 128 Hz), from 910,476 channel signals (about 65,000 events) in the original v1 database. Code -1 marks control signals captured without the digit stimulus. The author's 2022 Hugging Face release splits the data 80/20 into train.csv and test.csv (stratified over the 11 classes), resamples to the headset's nominal rate and adds headers named ChannelName-SampleNum, so each row is one labelled event; about 2.8 GB in total. Load with pandas or polars for EEG digit classification, brain-computer interface (BCI) and consumer-grade EEG benchmarking. Stated on mindbigdata.com: the MNIST of Brain Digits is made available under the Open Database License (ODbL 1.0), with contents under the Database Contents License. Cite Vivancos and Cuesta, MindBigData 2022: A Large Dataset of Brain Signals, arXiv:2212.14746.

## Load the data

```python
import pandas as pd
df = pd.read_csv("hf://datasets/DavidVivancos/MindBigData2022_MNIST_EP/test.csv")
df.head()
```

## Data structure

File types in sampled files: CSV

Example file: `test.csv`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/6e56b957-a7a9-536f-b4d0-7b0571b5703b
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/6e56b957-a7a9-536f-b4d0-7b0571b5703b
- Over MCP: `get_schema_card` with id `6e56b957-a7a9-536f-b4d0-7b0571b5703b` at https://api.scigantic.com/mcp

## Related topics

- [Neuroscience datasets](https://scigantic.com/topics/neuroscience)
- [CSV datasets](https://scigantic.com/topics/csv)
- [EEG datasets](https://scigantic.com/topics/eeg)
