README.md

metadata

language:
  - en
size_categories:
  - 1B<n<10B
source_datasets:
  - database.lichess.org
task_categories:
  - other
paperswithcode_id: lichess_uci
pretty_name: Lichess.org Database in UCI format
tags:
  - chess
  - UCI
  - Lichess
dataset_info:
  - config_name: 201301-headers
    features:
      - name: event
        dtype: string
      - name: site
        dtype: string
      - name: white
        dtype: string
      - name: black
        dtype: string
      - name: result
        dtype: string
      - name: utcdate
        dtype: date32
      - name: utctime
        dtype: time64[us]
      - name: whiteelo
        dtype: int64
      - name: blackelo
        dtype: int64
      - name: whiteratingdiff
        dtype: float64
      - name: blackratingdiff
        dtype: float64
      - name: eco
        dtype: string
      - name: opening
        dtype: string
      - name: timecontrol
        dtype: string
      - name: termination
        dtype: string
    splits:
      - name: train
        num_bytes: 20700133
        num_examples: 121332
    download_size: 6335452
    dataset_size: 20700133
  - config_name: 201301-moves
    features:
      - name: site
        dtype: string
      - name: transcript
        dtype: string
    splits:
      - name: train
        num_bytes: 42454117.656856485
        num_examples: 120133
    download_size: 992204415449
    dataset_size: 1636203333850.0193
  - config_name: 201302-headers
    features:
      - name: event
        dtype: string
      - name: site
        dtype: string
      - name: white
        dtype: string
      - name: black
        dtype: string
      - name: result
        dtype: string
      - name: utcdate
        dtype: date32
      - name: utctime
        dtype: time64[us]
      - name: whiteelo
        dtype: int64
      - name: blackelo
        dtype: int64
      - name: whiteratingdiff
        dtype: float64
      - name: blackratingdiff
        dtype: float64
      - name: eco
        dtype: string
      - name: opening
        dtype: string
      - name: timecontrol
        dtype: string
      - name: termination
        dtype: string
    splits:
      - name: train
        num_bytes: 21155456
        num_examples: 123961
    download_size: 6487839
    dataset_size: 21155456
  - config_name: 201303-headers
    features:
      - name: event
        dtype: string
      - name: site
        dtype: string
      - name: white
        dtype: string
      - name: black
        dtype: string
      - name: result
        dtype: string
      - name: utcdate
        dtype: date32
      - name: utctime
        dtype: time64[us]
      - name: whiteelo
        dtype: int64
      - name: blackelo
        dtype: int64
      - name: whiteratingdiff
        dtype: float64
      - name: blackratingdiff
        dtype: float64
      - name: eco
        dtype: string
      - name: opening
        dtype: string
      - name: timecontrol
        dtype: string
      - name: termination
        dtype: string
    splits:
      - name: train
        num_bytes: 27026323
        num_examples: 158635
    download_size: 8300668
    dataset_size: 27026323
  - config_name: 201304-headers
    features:
      - name: event
        dtype: string
      - name: site
        dtype: string
      - name: white
        dtype: string
      - name: black
        dtype: string
      - name: result
        dtype: string
      - name: utcdate
        dtype: date32
      - name: utctime
        dtype: time64[us]
      - name: whiteelo
        dtype: int64
      - name: blackelo
        dtype: int64
      - name: whiteratingdiff
        dtype: float64
      - name: blackratingdiff
        dtype: float64
      - name: eco
        dtype: string
      - name: opening
        dtype: string
      - name: timecontrol
        dtype: string
      - name: termination
        dtype: string
    splits:
      - name: train
        num_bytes: 27075305
        num_examples: 157871
    download_size: 8357025
    dataset_size: 27075305
  - config_name: 201305-headers
    features:
      - name: event
        dtype: string
      - name: site
        dtype: string
      - name: white
        dtype: string
      - name: black
        dtype: string
      - name: result
        dtype: string
      - name: utcdate
        dtype: date32
      - name: utctime
        dtype: time64[us]
      - name: whiteelo
        dtype: int64
      - name: blackelo
        dtype: int64
      - name: whiteratingdiff
        dtype: float64
      - name: blackratingdiff
        dtype: float64
      - name: eco
        dtype: string
      - name: opening
        dtype: string
      - name: timecontrol
        dtype: string
      - name: termination
        dtype: string
    splits:
      - name: train
        num_bytes: 30787801
        num_examples: 179550
    download_size: 9499827
    dataset_size: 30787801
  - config_name: 201306-headers
    features:
      - name: event
        dtype: string
      - name: site
        dtype: string
      - name: white
        dtype: string
      - name: black
        dtype: string
      - name: result
        dtype: string
      - name: utcdate
        dtype: date32
      - name: utctime
        dtype: time64[us]
      - name: whiteelo
        dtype: int64
      - name: blackelo
        dtype: int64
      - name: whiteratingdiff
        dtype: float64
      - name: blackratingdiff
        dtype: float64
      - name: eco
        dtype: string
      - name: opening
        dtype: string
      - name: timecontrol
        dtype: string
      - name: termination
        dtype: string
    splits:
      - name: train
        num_bytes: 38460608
        num_examples: 224679
    download_size: 12011278
    dataset_size: 38460608
description: >-
  "The dataset contains games from the Lichess.org open database converted from
  PGN format to UCI format. It is divided into two main configurations: `moves`
  and `headers`. The `moves` configuration includes a UUID called `site` and the
  UCI moves in a column called `transcript`, while the `headers` configuration
  includes metadata such as player ELO ratings, game outcome, and dates. The
  data is subset based on the year and month (yyyymm format)." 
configs:
  - config_name: 201301-headers
    data_files:
      - split: train
        path: headers/201301/train-*
  - config_name: 201301-moves
    data_files:
      - split: train
        path: data/201301-*
    default: true
  - config_name: 201302-headers
    data_files:
      - split: train
        path: headers/201302/train-*
  - config_name: 201302-moves
    data_files:
      - split: train
        path: data/201302-*
  - config_name: 201303-headers
    data_files:
      - split: train
        path: headers/201303/train-*
  - config_name: 201303-moves
    data_files:
      - split: train
        path: data/201303-*
  - config_name: 201304-headers
    data_files:
      - split: train
        path: headers/201304/train-*
  - config_name: 201304-moves
    data_files:
      - split: train
        path: data/201304-*
  - config_name: 201305-headers
    data_files:
      - split: train
        path: headers/201305/train-*
  - config_name: 201305-moves
    data_files:
      - split: train
        path: data/201305-*
  - config_name: 201306-headers
    data_files:
      - split: train
        path: headers/201306/train-*
  - config_name: 201306-moves
    data_files:
      - split: train
        path: data/201306-*
  - config_name: 201307-moves
    data_files:
      - split: train
        path: data/201307-*
  - config_name: 201308-moves
    data_files:
      - split: train
        path: data/201308-*
  - config_name: 201309-moves
    data_files:
      - split: train
        path: data/201309-*
  - config_name: 201310-moves
    data_files:
      - split: train
        path: data/201310-*
  - config_name: 201311-moves
    data_files:
      - split: train
        path: data/201311-*
  - config_name: 201312-moves
    data_files:
      - split: train
        path: data/201312-*
  - config_name: 201401-moves
    data_files:
      - split: train
        path: data/201401-*
  - config_name: 201402-moves
    data_files:
      - split: train
        path: data/201402-*
  - config_name: 201403-moves
    data_files:
      - split: train
        path: data/201403-*
  - config_name: 201404-moves
    data_files:
      - split: train
        path: data/201404-*
  - config_name: 201405-moves
    data_files:
      - split: train
        path: data/201405-*
  - config_name: 201406-moves
    data_files:
      - split: train
        path: data/201406-*
  - config_name: 201407-moves
    data_files:
      - split: train
        path: data/201407-*
  - config_name: 201408-moves
    data_files:
      - split: train
        path: data/201408-*
  - config_name: 201409-moves
    data_files:
      - split: train
        path: data/201409-*
  - config_name: 201410-moves
    data_files:
      - split: train
        path: data/201410-*
  - config_name: 201411-moves
    data_files:
      - split: train
        path: data/201411-*
  - config_name: 201412-moves
    data_files:
      - split: train
        path: data/201412-*
  - config_name: 201501-moves
    data_files:
      - split: train
        path: data/201501-*
  - config_name: 201502-moves
    data_files:
      - split: train
        path: data/201502-*
  - config_name: 201503-moves
    data_files:
      - split: train
        path: data/201503-*
  - config_name: 201504-moves
    data_files:
      - split: train
        path: data/201504-*
  - config_name: 201505-moves
    data_files:
      - split: train
        path: data/201505-*
  - config_name: 201506-moves
    data_files:
      - split: train
        path: data/201506-*
  - config_name: 201507-moves
    data_files:
      - split: train
        path: data/201507-*
  - config_name: 201508-moves
    data_files:
      - split: train
        path: data/201508-*
  - config_name: 201509-moves
    data_files:
      - split: train
        path: data/201509-*
  - config_name: 201510-moves
    data_files:
      - split: train
        path: data/201510-*
  - config_name: 201511-moves
    data_files:
      - split: train
        path: data/201511-*
  - config_name: 201512-moves
    data_files:
      - split: train
        path: data/201512-*
  - config_name: 201601-moves
    data_files:
      - split: train
        path: data/201601-*
  - config_name: 201602-moves
    data_files:
      - split: train
        path: data/201602-*
  - config_name: 201603-moves
    data_files:
      - split: train
        path: data/201603-*
  - config_name: 201604-moves
    data_files:
      - split: train
        path: data/201604-*
  - config_name: 201605-moves
    data_files:
      - split: train
        path: data/201605-*
  - config_name: 201606-moves
    data_files:
      - split: train
        path: data/201606-*
  - config_name: 201607-moves
    data_files:
      - split: train
        path: data/201607-*
  - config_name: 201608-moves
    data_files:
      - split: train
        path: data/201608-*
  - config_name: 201609-moves
    data_files:
      - split: train
        path: data/201609-*
  - config_name: 201610-moves
    data_files:
      - split: train
        path: data/201610-*
  - config_name: 201611-moves
    data_files:
      - split: train
        path: data/201611-*
  - config_name: 201612-moves
    data_files:
      - split: train
        path: data/201612-*
  - config_name: 201701-moves
    data_files:
      - split: train
        path: data/201701-*
  - config_name: 201702-moves
    data_files:
      - split: train
        path: data/201702-*
  - config_name: 201703-moves
    data_files:
      - split: train
        path: data/201703-*
  - config_name: 201704-moves
    data_files:
      - split: train
        path: data/201704-*
  - config_name: 201705-moves
    data_files:
      - split: train
        path: data/201705-*
  - config_name: 201706-moves
    data_files:
      - split: train
        path: data/201706-*
  - config_name: 201707-moves
    data_files:
      - split: train
        path: data/201707-*
  - config_name: 201708-moves
    data_files:
      - split: train
        path: data/201708-*
  - config_name: 201709-moves
    data_files:
      - split: train
        path: data/201709-*
  - config_name: 201710-moves
    data_files:
      - split: train
        path: data/201710-*
  - config_name: 201711-moves
    data_files:
      - split: train
        path: data/201711-*
  - config_name: 201712-moves
    data_files:
      - split: train
        path: data/201712-*
  - config_name: 201801-moves
    data_files:
      - split: train
        path: data/201801-*
  - config_name: 201802-moves
    data_files:
      - split: train
        path: data/201802-*
  - config_name: 201803-moves
    data_files:
      - split: train
        path: data/201803-*
  - config_name: 201804-moves
    data_files:
      - split: train
        path: data/201804-*
  - config_name: 201805-moves
    data_files:
      - split: train
        path: data/201805-*
  - config_name: 201806-moves
    data_files:
      - split: train
        path: data/201806-*
  - config_name: 201807-moves
    data_files:
      - split: train
        path: data/201807-*
  - config_name: 201808-moves
    data_files:
      - split: train
        path: data/201808-*
  - config_name: 201809-moves
    data_files:
      - split: train
        path: data/201809-*
  - config_name: 201810-moves
    data_files:
      - split: train
        path: data/201810-*
  - config_name: 201811-moves
    data_files:
      - split: train
        path: data/201811-*
  - config_name: 201812-moves
    data_files:
      - split: train
        path: data/201812-*
  - config_name: 201901-moves
    data_files:
      - split: train
        path: data/201901-*
  - config_name: 201902-moves
    data_files:
      - split: train
        path: data/201902-*
  - config_name: 201903-moves
    data_files:
      - split: train
        path: data/201903-*
  - config_name: 201904-moves
    data_files:
      - split: train
        path: data/201904-*
  - config_name: 201905-moves
    data_files:
      - split: train
        path: data/201905-*
  - config_name: 201906-moves
    data_files:
      - split: train
        path: data/201906-*
  - config_name: 201907-moves
    data_files:
      - split: train
        path: data/201907-*
  - config_name: 201908-moves
    data_files:
      - split: train
        path: data/201908-*
  - config_name: 201909-moves
    data_files:
      - split: train
        path: data/201909-*
  - config_name: 201910-moves
    data_files:
      - split: train
        path: data/201910-*
  - config_name: 201911-moves
    data_files:
      - split: train
        path: data/201911-*
  - config_name: 201912-moves
    data_files:
      - split: train
        path: data/201912-*
  - config_name: 202001-moves
    data_files:
      - split: train
        path: data/202001-*
  - config_name: 202002-moves
    data_files:
      - split: train
        path: data/202002-*
  - config_name: 202003-moves
    data_files:
      - split: train
        path: data/202003-*
  - config_name: 202004-moves
    data_files:
      - split: train
        path: data/202004-*
  - config_name: 202005-moves
    data_files:
      - split: train
        path: data/202005-*
  - config_name: 202006-moves
    data_files:
      - split: train
        path: data/202006-*
  - config_name: 202007-moves
    data_files:
      - split: train
        path: data/202007-*
  - config_name: 202008-moves
    data_files:
      - split: train
        path: data/202008-*
  - config_name: 202009-moves
    data_files:
      - split: train
        path: data/202009-*
  - config_name: 202010-moves
    data_files:
      - split: train
        path: data/202010-*
  - config_name: 202011-moves
    data_files:
      - split: train
        path: data/202011-*
  - config_name: 202012-moves
    data_files:
      - split: train
        path: data/202012-*
  - config_name: 202101-moves
    data_files:
      - split: train
        path: data/202101-*
  - config_name: 202102-moves
    data_files:
      - split: train
        path: data/202102-*
  - config_name: 202103-moves
    data_files:
      - split: train
        path: data/202103-*
  - config_name: 202104-moves
    data_files:
      - split: train
        path: data/202104-*
  - config_name: 202105-moves
    data_files:
      - split: train
        path: data/202105-*
  - config_name: 202106-moves
    data_files:
      - split: train
        path: data/202106-*
  - config_name: 202107-moves
    data_files:
      - split: train
        path: data/202107-*
  - config_name: 202108-moves
    data_files:
      - split: train
        path: data/202108-*
  - config_name: 202109-moves
    data_files:
      - split: train
        path: data/202109-*
  - config_name: 202110-moves
    data_files:
      - split: train
        path: data/202110-*
  - config_name: 202111-moves
    data_files:
      - split: train
        path: data/202111-*
  - config_name: 202112-moves
    data_files:
      - split: train
        path: data/202112-*
  - config_name: 202201-moves
    data_files:
      - split: train
        path: data/202201-*
  - config_name: 202202-moves
    data_files:
      - split: train
        path: data/202202-*
  - config_name: 202203-moves
    data_files:
      - split: train
        path: data/202203-*
  - config_name: 202204-moves
    data_files:
      - split: train
        path: data/202204-*
  - config_name: 202205-moves
    data_files:
      - split: train
        path: data/202205-*
  - config_name: 202206-moves
    data_files:
      - split: train
        path: data/202206-*
  - config_name: 202207-moves
    data_files:
      - split: train
        path: data/202207-*
  - config_name: 202208-moves
    data_files:
      - split: train
        path: data/202208-*
  - config_name: 202209-moves
    data_files:
      - split: train
        path: data/202209-*
  - config_name: 202210-moves
    data_files:
      - split: train
        path: data/202210-*
  - config_name: 202211-moves
    data_files:
      - split: train
        path: data/202211-*
  - config_name: 202212-moves
    data_files:
      - split: train
        path: data/202212-*
  - config_name: 202301-moves
    data_files:
      - split: train
        path: data/202301-*
  - config_name: 202302-moves
    data_files:
      - split: train
        path: data/202302-*
  - config_name: 202303-moves
    data_files:
      - split: train
        path: data/202303-*
  - config_name: 202304-moves
    data_files:
      - split: train
        path: data/202304-*
  - config_name: 202305-moves
    data_files:
      - split: train
        path: data/202305-*
  - config_name: 202306-moves
    data_files:
      - split: train
        path: data/202306-*
  - config_name: 202307-moves
    data_files:
      - split: train
        path: data/202307-*
  - config_name: 202308-moves
    data_files:
      - split: train
        path: data/202308-*
  - config_name: 202309-moves
    data_files:
      - split: train
        path: data/202309-*
  - config_name: 202310-moves
    data_files:
      - split: train
        path: data/202310-*
  - config_name: 202311-moves
    data_files:
      - split: train
        path: data/202311-*
  - config_name: 202312-moves
    data_files:
      - split: train
        path: data/202312-*
  - config_name: 202401-moves
    data_files:
      - split: train
        path: data/202401-*
  - config_name: 202402-moves
    data_files:
      - split: train
        path: data/202402-*
example:
  moves:
    site: xxxxxxx
    transcript: e2e4 e7e5 g1f3 b8c6
  headers:
    event: Rated Blitz game
    site: xxxxxxx
    white: player1
    black: player2
    result: 1-0
    utcdate: '2023-05-21'
    utctime: '13:45:00'
    whiteelo: 1500
    blackelo: 1400
    whiteratingdiff: 10
    blackratingdiff: -10
    eco: C50
    opening: Italian Game
    timecontrol: 300+0
    termination: Normal

Dataset Card for Lichess.org Database in UCI format

Table of Contents

• Table of Contents
• Dataset Description
  • Dataset Summary
  • Supported Tasks and Leaderboards
• Dataset Structure
  • Data Instances
  • Data Fields
  • Data Splits
• Dataset Creation
  • Curation Rationale
  • Source Data
  • Annotations
  • Personal and Sensitive Information
• Considerations for Using the Data
  • Social Impact of Dataset
  • Discussion of Biases
  • Other Known Limitations
• Additional Information
  • Dataset Curators
  • Licensing Information
  • Citation Information
  • Contributions

Dataset Description

Dataset Summary

The Lichess.org database has been converted to UCI format, making it easier to analyze and interpret chess games. The moves configuration captures the sequence of moves using UCI notation, and the headers configuration provides comprehensive metadata for each game, enabling detailed statistical and strategic analysis. The data is subset based on the year and month (yyyymm format) the games took place.

Supported Tasks and Leaderboards

This dataset supports tasks related to chess game analysis, including move prediction, game outcome prediction, performance analysis, and opening strategy evaluation. There are no formal leaderboards associated with this dataset.

Dataset Structure

Data Instances

An example from the moves configuration:

{
  "site": "abcd1234",
  "transcript": "e2e4 e7e5 g1f3 b8c6"
}

An example from the headers configuration:

{
  "event": "Rated Blitz game",
  "site": "abcd1234",
  "white": "player1",
  "black": "player2",
  "result": "1-0",
  "utcdate": "2023-05-21",
  "utctime": "13:45:00",
  "whiteelo": 1500,
  "blackelo": 1400,
  "whiteratingdiff": 10,
  "blackratingdiff": -10,
  "eco": "C50",
  "opening": "Italian Game",
  "timecontrol": "300+0",
  "termination": "Normal"
}

Data Fields

Moves Configuration:

• site: string, unique identifier for the game. Replays can be viewed by navigating to https://lichess.org/<site>
• transcript: string, sequence of moves in UCI format.

Headers Configuration:

• event: string, type of event.
• site: string, unique identifier for the game.
• white: string, white player.
• black: string, black player.
• result: string, game result.
• utcdate: date32, date of the game.
• utctime: time64[us], time of the game.
• whiteelo: int64, ELO rating of the white player.
• blackelo: int64, ELO rating of the black player.
• whiteratingdiff: float64, rating change for the white player.
• blackratingdiff: float64, rating change for the black player.
• eco: string, ECO code of the opening.
• opening: string, name of the opening.
• timecontrol: string, time control format.
• termination: string, reason for game termination.

Data Splits

The dataset is divided into monthly splits based on the year and month (yyyymm format). Each split contains both moves and headers configurations.

Dataset Creation

Curation Rationale

The dataset was curated to facilitate research and analysis that use chess games, providing both move sequences and comprehensive metadata.

Source Data

Initial Data Collection and Normalization

The data was collected from the Lichess.org open database, converted from PGN format to UCI format, and organized into moves and headers configurations.

Who are the source data producers?

The source data comes from games played on Lichess.org, an online platform where users from around the world play chess. This database does not distinguish between human-and bot-played games. However, it's reasonable to assume games played before April 2018 were overwhelmingly played by human players since Lichess.org released its bot api on April 1st, 2018.

Annotations

Annotation process

The annotations include metadata such as player ELO ratings, game outcomes, and dates, which were extracted from the original PGN files and normalized into a structured format.

Who are the annotators?

The annotations were generated by the Lichess.org platform and curated by the dataset creators.

Personal and Sensitive Information

The dataset does not contain any personal or sensitive information.

Considerations for Using the Data

Social Impact of Dataset

The dataset can be used to improve chess engines, develop training tools for players, and conduct research in game theory and artificial intelligence.

Discussion of Biases

The dataset may have biases related to the demographics of Lichess.org users, such as skill level distribution and regional representation.

Other Known Limitations

The dataset is limited to games played on Lichess.org and may not represent the broader chess-playing population.

Additional Information

Dataset Curators

The dataset was curated by the Lichess.org team and contributors.

Licensing Information

The dataset is available under the Creative Commons Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license.

Citation Information

If you use this dataset, please cite it as follows:

@misc{lichess_uci,
  author = {Davis, Austin L.},
  title = {Lichess.org Database in UCI format},
  year = {2023},
  howpublished = {\url{https://database.lichess.org/}},
}

Contributions

Thanks to @github-username for adding this dataset.