IMBd-dataset

SOURCE: https://ai.stanford.edu/~amaas/data/sentiment/

The files in ´./raw-data' have been extracted directly from the source.

Using duckdb:

SELECT * FROM 'https://github.com/Digital-Defiance/IMBd-dataset/raw/main/dataset/train.parquet';
SELECT * FROM 'https://github.com/Digital-Defiance/IMBd-dataset/raw/main/dataset/test.parquet';

To reproduce this dataset:

rm -rf dataset
pip install -r requirements.txt
python pipeline.py