Outline the dataset structure

README.md

@@ -1,3 +1,59 @@
 ---
 license: agpl-3.0
+configs:
+- config_name: asciiart
+  data_files:
+  - split: train
+    path: "asciiart/train/*.parquet"
+- config_name: copypasta
+  default: true
+  data_files:
+  - split: train
+    path: "copypasta/train/*.parquet"
+- config_name: graffiti
+  default: true
+  data_files:
+  - split: train
+    path: "graffiti/train/*.parquet"
+- config_name: images
+  default: true
+  data_files:
+  - split: train
+    path: "images/train/*.parquet"
 ---
+
+# ASCII Art
+
+## Description
+
+Collection of ASCII art from various sources:
+
+- `asciiart`: made by independent artists, listed on [asciiart.eu][source-asciiart]
+- `copypasta`: common twitch emotes, listed on [twitchquotes.com][source-twitchquotes]
+- `graffiti`: text samples styled using various ASCII art fonts (soon)
+- `images`: image conversion using tools, on text-to-image datasets (soon)
+
+## Metadata
+
+- homepage: [https://github.com/apehex/scrapscii][github-scrapscii]
+- version: 0.2.0
+
+| Config            | Split     | Size      | Samples   |
+| ----------------- | --------- | --------- | --------- |
+| 'asciiart'        | 'train'   | 1.8 MB    | 5293      |
+| 'copypasta'       | 'train'   | 488 KB    | 909       |
+
+## Features
+
+```python
+datasets.Features({
+    'caption': datasets.features.Value(dtype='uint64'),
+    'content': datasets.features.Value(dtype='uint64'),
+    'labels': datasets.features.Value(dtype='large_binary'),
+    'charsets': datasets.features.Value(dtype='large_binary'),
+    'chartypes': datasets.features.Value(dtype='large_binary'),})
+```
+
+[github-scrapscii]: https://github.com/apehex/scrapscii/
+[source-asciiart]: https://www.asciiart.eu/
+[source-twitchquotes]: https://www.twitchquotes.com/