Datasets:
Guillaume Raille
commited on
add readme
Browse files
README.md
CHANGED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# tiny-moliere
|
| 2 |
+
|
| 3 |
+
A dataset repo generating `tinymoliere.txt` containing Moli�re's complete work.
|
| 4 |
+
|
| 5 |
+
Inspired by [tinyshakespeare](https://github.com/karpathy/char-rnn/blob/master/data/tinyshakespeare/input.txt) by Andrej Karpathy, this project provides a consolidated small text corpus ideal for training and learning with small transformer models.
|
| 6 |
+
|
| 7 |
+
## What it does
|
| 8 |
+
|
| 9 |
+
Downloads Moli�re'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.
|
| 10 |
+
|
| 11 |
+
## Usage
|
| 12 |
+
|
| 13 |
+
```bash
|
| 14 |
+
uv sync
|
| 15 |
+
uv run python main.py
|
| 16 |
+
```
|
| 17 |
+
|
| 18 |
+
This will:
|
| 19 |
+
1. Download the source PDFs to `data/` directory
|
| 20 |
+
2. Process and clean the text
|
| 21 |
+
3. Generate `data/tinymoliere.txt`
|