Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,84 @@
|
|
1 |
-
---
|
2 |
-
license: cc0-1.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc0-1.0
|
3 |
+
tags:
|
4 |
+
- chess
|
5 |
+
- lichess
|
6 |
+
- game
|
7 |
+
- games
|
8 |
+
pretty_name: Antichess Openings
|
9 |
+
---
|
10 |
+
|
11 |
+
# Dataset Card for Lichess Puzzles
|
12 |
+
|
13 |
+
<!-- Provide a quick summary of the dataset. -->
|
14 |
+
## Dataset Description
|
15 |
+
|
16 |
+
A list of Antichess openings.
|
17 |
+
|
18 |
+
### Dataset Creation
|
19 |
+
|
20 |
+
This dataset was compiled by **[tolius](https://lichess.org/@/tolius)**.
|
21 |
+
|
22 |
+
### Dataset Usage
|
23 |
+
|
24 |
+
Using the `datasets` library:
|
25 |
+
|
26 |
+
```python
|
27 |
+
from datasets import load_dataset
|
28 |
+
|
29 |
+
dset = load_dataset("Lichess/antichess-chess-openings", split="train")
|
30 |
+
```
|
31 |
+
Using the `pandas` library:
|
32 |
+
|
33 |
+
```python
|
34 |
+
import pandas as pd
|
35 |
+
|
36 |
+
df = pd.read_parquet("hf://datasets/Lichess/antichess-chess-openings/data/train-00000-of-00001.parquet")
|
37 |
+
```
|
38 |
+
|
39 |
+
Using the `polars` library:
|
40 |
+
|
41 |
+
```python
|
42 |
+
import polars as pl
|
43 |
+
|
44 |
+
df = pl.read_parquet('hf://datasets/Lichess/chess-openings/data/train-00000-of-00001.parquet')
|
45 |
+
```
|
46 |
+
|
47 |
+
## Dataset Details
|
48 |
+
|
49 |
+
### Dataset Sample
|
50 |
+
|
51 |
+
One row of the dataset looks like this:
|
52 |
+
|
53 |
+
```python
|
54 |
+
{
|
55 |
+
'name':
|
56 |
+
'pgn':
|
57 |
+
'uci':
|
58 |
+
'epd':
|
59 |
+
}
|
60 |
+
```
|
61 |
+
### Dataset Fields
|
62 |
+
|
63 |
+
Every row of the dataset contains the following fields:
|
64 |
+
|
65 |
+
- **`name`**: `string`, the name of the opening in English.
|
66 |
+
- **`pgn`**: `int`, the sequence of moves leading to the opening position. If not unique, the moves are the most common ones leading to that opening.
|
67 |
+
- **`uci`**: `int`, the same moves described by the `pgn` field in [UCI notation](https://backscattering.de/chess/uci/#move).
|
68 |
+
- **`epd`**: `int`, the [EPD](https://www.chessprogramming.org/Extended_Position_Description) ([FEN](https://en.wikipedia.org/wiki/Forsyth%E2%80%93Edwards_Notation) without move numbers) of the opening position, en passant field only if legal
|
69 |
+
|
70 |
+
### Dataset Conventions
|
71 |
+
|
72 |
+
* Title case is used for opening names.
|
73 |
+
* Names are structured like `Opening family: Variation, Subvariation, ...`,
|
74 |
+
e.g., `Sicilian Defense: Najdorf Variation, English Attack`.
|
75 |
+
* The suggested way to classify games is to play moves backwards until
|
76 |
+
a named position is found. To make this work well with common transpositions,
|
77 |
+
multiple entries for a single opening may be added.
|
78 |
+
* However, each name has a unique *shortest* line. If necessary,
|
79 |
+
a distinguishing move is appended, e.g.,
|
80 |
+
`King's Gambit Accepted: Schurig Gambit, with Bb5`.
|
81 |
+
|
82 |
+
## Additional Information
|
83 |
+
|
84 |
+
- To contribute to the dataset, send a PR or open an issue in the community tab
|