metadata

language:
  - fr
license: mit
size_categories:
  - 1M<n<10M
task_categories:
  - text-generation
pretty_name: Tiny Molière
dataset_info:
  features:
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 2387600
      num_examples: 1
  download_size: 2387600
  dataset_size: 2387600
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/tinymoliere.txt
tags:
  - literature
  - french-literature
  - moliere
  - classical-text
  - character-level

tiny-moliere

A dataset repo generating tinymoliere.txt containing Molière's complete work.

Inspired by tinyshakespeare by Andrej Karpathy, this project provides a consolidated small text corpus ideal for training and learning with small transformer models.

What it does

Downloads Molière's complete works from public PDFs, processes them to remove headers/footers and table of contents, then outputs a single clean text file suitable for machine learning tasks.

Usage

uv sync
uv run python main.py

This will:

Download the source PDFs to data/ directory
Process and clean the text
Generate data/tinymoliere.txt