images

"Halli Galli" is a tabletop game centered around quick reactions. The game consists of 54 cards featuring 4 types of fruit: bananas, lemons, strawberries, and grapes. Each card shows between 1 and 5 fruits. The main mechanism of the game is for players to take turns playing cards from their hand. When a fruit appears five times or in multiples of five, the first player to notice and ring the bell wins all the cards on the table, placing them face down in their pile. If a player rings the bell incorrectly, they must give each player one card as a penalty. The game continues until a player runs out of cards, at which point they are eliminated. https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F19542212%2F3ec6d4be646b01c0a5d269d473baf64a%2F3.jpg?generation=1710233513066528&alt=media" alt=""> https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F19542212%2F99e3a3c8a82cfed237eb6bff490985f3%2F20.jpg?generation=1710233543699702&alt=media" alt=""> We created a dataset for the board game "Halli Galli" to simulate different random card-playing scenarios. The dataset involves randomly placing the 54 cards on a fixed-size tabletop, generating data, and determining the wins and losses of the images based on the game rules. When the quantity of a certain fruit is 5 or a multiple of 5, the image label is 1; otherwise, it is 0. The dataset contains 20,000 training images, 2,000 validation images, and 2,000 test images.This dataset aims to explore the semantic understanding and logical reasoning abilities of various visual models in the absence of given game rules. We hope to discover a visual model with logical reasoning capabilities through this dataset, providing a new direction for development in the field of computer vision.If you want to know about the data generation code and the related models' performance on this dataset, please visit my repository:https://github.com/gitcat-404/Halli-Galli-Dataset

Between January 1st and September 3rd 2024, the most played game by users of BoardGameGeek, an online forum and database for board gaming hobbyists, was Ark Nova with more than 100,000 recorded plays. Meanwhile, Azul and Wingspan ranked second and third with 79,247 and 73,970 games played respectively.

As of September 2024, the highest rated game by users of BoardGameGeek, an online forum and database for board gaming hobbyists, was Brass: Birmingham with an average user rating of 8.41 out of 10. It is followed in the rankings by Pandemic Legacy: Season 1 and Gloomhaven.

Context

Dataset was made from 17 professional shogi players' Youtube channels with Youtube Data API. I made a dataset from one of the channels before with Selenium on https://www.kaggle.com/datasets/satoshiss/shogi-channels-data.

If you are interested in Shogi(Japanese Chess), please check any videos listed.

Content

The channel stats file has overall stats for each youtube channel and the video_details file have information on each video including title, views, likes, comment counts, tags, description, and published date.

Description:

👉 Download the dataset here

Ticket to Ride Games is a popular strategic board game where players compete to connect various cities on a map by placing their train pieces along specific routes. The objective is to complete the longest and most valuable routes while blocking opponents' paths. The game involves strategic planning, as each player's moves impact the overall board configuration and the availability of routes. This dataset is designed for enthusiasts, researchers, and Al developers interested in analyzing board game strategies, computer vision tasks, and data-driven game mechanics. It provides a comprehensive look at how different players approach the game, offering insights into decision-making processes, route optimization, and game dynamics.

Download Dataset

Content:

The dataset contains high-resolution images capturing various board configurations during the game, showcasing the players' city connections. The images are taken from four different angles to provide a complete view of the board's layout. Additionally, the dataset includes a well-structure CSV file that labels each player's city connections, detailing which cities have been successfully link by train routes. This labeling allows for in-depth analysis and pattern recognition, making it an ideal resource for those interest in game theory, Al training, or visual recognition models.

Key Features:

Images: Over [insert number] high-quality images of board configurations during gameplay, capturing different stages and strategies employe by players.

Angles: Each configuration is photograph from four distinct angles, ensuring comprehensive visual data for analysis.

CSV Labeling: The accompanying CSV file provides detail labeling of players' city connections, specifying which routes have been claim by each player. This structured data enables various analytical approaches, including statistical analysis, machine learning, and Al model training.

Versatile Applications: The dataset can be use for computer vision tasks, such as object detection and image segmentation, as well as for developing Al models to simulate or predict player strategies in board games.

Research Potential: Ideal for academic research, game development, and Al training, this dataset offers a rich source of data for exploring the complexities of board game strategies and player behaviors.

This dataset is sourced from Kaggle.

Here are a few use cases for this project:

1. Board Game Assistants: This Dice model can be used in a digital assistant for board games. It would help users track dice results automatically, thereby enhancing the experience for games involving dice such as Monopoly, Yahtzee, or Dungeons and Dragons.

2. Educational Games Development: Educational organizations and ed-tech companies can use this model to develop interactive learning games or applications that teach probability, math or statistics through a dice game.

3. Gambling Supervision: Casinos or online gambling platforms can apply the model to monitor dice games and ensure fair play, automatically and meticulously track game statistics, and verify or dispute any contentious throws.

4. Virtual Reality Gaming: The Dice model can be integrated into VR gaming systems to interact with physical dice. For instance, in a VR board game setup, the model can compute the numbers rolled on the dice and translate that into the virtual game.

5. Assistive Technology for Visually-Impaired: Application for visually impaired people, where the app can detect the number rolled on a dice and communicate it via audio, enabling visually impaired people to participate in dice-based games.

Context

I scraped data from boardgaming-online.com, in order to analyze it and improve my skill at the board game "Through the Ages".

Content

The raw data consists of over 30k game journals scraped from the website. Each of the game journals is in the form of a pandas DataFrame serialized to a file as a pickle (a type of object store built into Python). These are uploaded in a ZIP file and available from ../input/tta/TTA/ folder when using the dataset in a kernel.

There are also two cleaned and parsed files ready for further analysis. TechXYFullAlt was parsed with customized codes with my knowledge of the game, NLP34 is just a compressed version of the Journals, suitable for some Natural Language Processing. The Exploring the Data and some NLP kernel explores in some more details.

Acknowledgements

I thank the web-administrator for boardgameing-online for tolerating the scraping.

Inspiration

Well, like anyone else who plays a game: how do I win? :P

Here are a few use cases for this project:

1. Chess Game Analysis: Chess-I could be used to digitize physical games of chess and convert them into a digital format for later review, analysis or sharing. The model could also offer real-time insights about the position of the chess pieces.

2. Online Chess Platforms: The model could be implemented by online chess platforms to allow users to play physical games that are mirrored online. Players would simply use their own chess set at home, and the platform would utilize Chess-I to recognize the different chess pieces and their positions on the board, creating an immersive online-offline hybrid gaming experience.

3. Chess Tutoring and Learning Applications: Chess-I could be used in learning or tutoring applications to provide feedback on chess strategies, possible moves and their consequences. This would allow learners to play a physical game of chess while accessing informational assistance, improving their game.

4. Automated Chess Robots: In robotics, Chess-I could be used to power an automated chess-playing robot. The robot would use the vision model to identify the pieces on the board and decide on the next move, effectively playing a game of chess against a human opponent.

5. Broadcasting Physical Chess Matches: In sports broadcasting, Chess-I could be used to create real-time digital visualizations of important matches, making it easier for viewers to follow the game. This would be particularly useful in professional chess tournaments.

Here are a few use cases for this project:

1. Automated Chess Games Analysis: This model can be used to input and analyze games from physical chess boards in real-time or from photographs. It can transform these games into digital versions for further analysis, suggestions, or tracking player performance.

2. Developing Intelligent Chess Bots: The "Chess" model can aid in developing AI-based chess software/bots, which can recognize the state of the game by visually analyzing the chess board and decide the next move.

3. Assistive Technology for the Visually Impaired: It can be applied for developing applications assisting visually impaired people to play chess, by recognizing chess board state and using speech to tell the state or the possible moves.

4. Chess Learning and Coaching Apps: The model can be incorporated into educational applications that teach users how to play chess, understand chess movements, or improve their strategies. The app could provide real-time recommendations by recognizing the position of pieces on the board.

5. Cheat Detection in Competitive Chess: In competitive or online chess playing platforms, the model can be used to monitor games and detect anomalous activities or cheating attempts, such as piece position changes in a non-standard manner.

Here are a few use cases for this project:

1. Chess Game Analysis: This model could be used to analyze live or recorded chess games. It could automatically track the moves made by each player, enabling a detailed post-game review. This could be a valuable tool for chess players looking to improve their strategy.

2. Online Chess Platforms: The model could be integrated into an online chess platform. It could be used to create a feature where players could play a physical game of chess and the model would automatically update the game digitally based on the players' moves.

3. Chess Tutoring: The model could be used in an application designed to teach chess strategies and tactics. By identifying chess pieces and their positions, the app can guide students in learning about various chess strategies and scenarios.

4. Chess Robot Development: Robotics engineers could use the model while designing robots that can play chess. The model's ability to identify chess pieces could help the robot "understand" the game and decide on the best moves.

5. Board Game Digitization: The model could be used in a broader project to digitize various board games. The component that identifies the chess pieces could be a part of a larger system that allows users to play physical board games in a digital setting.

This is the dataset and model used for Tiny Towns Scorer, a computer vision project completed as part of CS 4664: Data-Centric Computing Capstone at Virginia Tech. The goal of the project was to calculate player scores in the board game Tiny Towns.

The dataset consists of 226 images and associated annotations, intended for object detection. The images are photographs of players' game boards over the course of a game of Tiny Towns, as well as photos of individual game pieces taken after the game. Photos were taken using hand-held smartphones. Images are in JPG and PNG formats. The annotations are provided in TFRecord 1.0 and CVAT for Images 1.1 formats.

The weights for the trained RetinaNet-portion of the model are also provided.

MahJong Board States Dataset

  Dataset Description

Dataset Name: MahJong Board States Dataset Summary: The MahJong Board States dataset contains an extensive collection of board states from Riichi Mahjong games, a popular variant of Mahjong in Japan. The dataset includes more than 650 million records collected from games played between 2009 and 2019. Each record describes the current state of the board and the actions of the players based on the hand and board… See the full description on the dataset page: https://huggingface.co/datasets/pjura/mahjong_board_states.

Here are a few use cases for this project:

1. Board Game Assistance: This computer vision model can be used to assist players during "Between Two Cities" card games. By identifying different card types, the model can offer suggestions for the next move, enhancing player strategy and improving engagement.

2. Digital Game Creation: Game developers can use the model during the creation of digital versions of this game. By interpreting the different card classes, it can be programmed to handle different gameplay scenarios accurately.

3. Online Game Streaming: Streamers can use the model for real-time card classification during live board game streams. This would allow audience members to understand the game better, as the system could provide simultaneous analysis and commentary.

4. Automated Game Scoring: The model can be used to develop an automatic scoring system during live gameplay. By identifying the cards, the system could calculate scores based on the rules of the game, reducing manual efforts.

5. Game Tutorial Creation: By using the model to identify cards, content developers can create interactive tutorials or demonstration videos for the game. The identified cards can provide context, making it easier for new players to understand the game rules and strategies.

Keep in Mind

It is important to mention that this dataset may not be suitable for all audiences, as it contains reviews that may include harsh language, offensive or toxic content, and ASCII art of inappropriate body parts. This might not be suitable for all users. We want to make it clear that we do not endorse or condone any of the content within the dataset. This information is presented solely as a means of providing an unfiltered and authentic view of how players experience CS:GO. Most of the time it's just trolling and shouldn't be taken too seriously, however, it is essential to acknowledge that the reviews included have not been censored in any way, shape or form - this is precisely how they were presented on the Steam website.

About the Dataset

This dataset contains a wealth of reviews for the highly acclaimed first-person shooter, CS:GO, or Counter Strike: Global Offensive.. Developed by Valve and Hidden Path Entertainment, the game's impressive longevity and continued player engagement is evident in the wide range of reviews included within this dataset. Featuring opinions on gameplay mechanics, graphics, overall game experience, and more, the dataset offers a vast array of perspectives from players across the board. The diverse mix of reviews lends itself to the possibility of a variety of use cases, including sentiment analysis, natural language processing, and machine learning. The inclusion of both positive and negative reviews ensures that the dataset is comprehensive, providing an accurate and detailed view of the sentiment surrounding the game. As such, this dataset offers valuable insights into the perception of CS:GO by its players and serves as an excellent resource for further research and analysis of the game's popularity, player satisfaction and overall experience.

Artwork source: https://www.artstation.com/artwork/vJyaZO

Here are a few use cases for this project:

1. Chess Game Analysis: Use the "chess pieces" computer vision model to analyze recorded chess games or live streaming matches to automatically track and record the moves made by each player. This data can be used to facilitate commentary, review games for improving player strategies, or generate GIFs and visualizations of game progress.

2. Online Chess Platforms: Integrate the computer vision model into online chess platforms and mobile apps to enable users to play chess by simply moving physical pieces on a real chessboard. The model can capture the positions of the pieces and update the digital board accordingly, allowing users to enjoy a more tactile and traditional experience while playing online.

3. Chess Tutoring: Use the "chess pieces" model to develop an AI-powered chess tutoring application that can analyze a student's movements on a physical board and provide real-time guidance, feedback, and suggestions for improvement. Such a tool can significantly engage and enhance the learning experience for beginners and intermediate players.

4. Augmented Reality Chess: The computer vision model can be employed to create augmented reality (AR) chess experiences for both entertainment and educational purposes. By identifying chess pieces and their positions on the board, the AR system can overlay annotations, suggested moves, and interactive elements on the physical board, making it more engaging for users.

5. Chess Content Curation: Implement the "chess pieces" model in content platforms or social media channels to automatically index and tag chess-related content for better categorization and discovery. Whether it's videos, articles, or images, the model can be used to accurately identify and classify content featuring chess elements, making it easier for users to find and consume relevant materials.

Oden Chess Dataset

  Dataset Summary

The Oden Chess Dataset is a comprehensive collection of over 4 million chess games compiled from top players, major tournaments, and categorized by opening systems. This dataset provides rich annotations including move sequences, board positions, player information, and game metadata, making it ideal for chess AI research, opening analysis, and statistical studies.

  Dataset Details

Total Games: 4,047,908 Source Files:… See the full description on the dataset page: https://huggingface.co/datasets/BBSRguy/Oden-worldchess.

ObjectiveDelay of gratification, or the extent to which one can resist the temptation of an immediate reward and wait for a larger reward later, is a self-regulatory skill that predicts positive outcomes. The aim of this research was to conduct initial tests of the effects of a board game designed to increase children’s delay of gratification via two experimental studies.MethodsPreschool children were randomized to play the study game or a control game. In Study 1, there were 48 children in the analytic sample, with a mean age of 4.81 ± 0.55 years; Study 2 included 50 children (M = 4.02 ± 0.76 years). Delay of gratification was assessed during the study game, as well as before and after game play sessions using the Marshmallow Test.ResultsIn both studies, the intervention group’s likelihood of delaying gratification during the study game increased across game-play sessions (p < 0.05). In Study 1, the intervention group also increased wait times during the Marshmallow Test versus controls (p = 0.047). In Study 2, there was no effect on Marshmallow Test wait times.ConclusionResults provide some initial evidence supporting potential efficacy of a board game designed to increase delay of gratification. Future research can clarify: (1) which components of game play (if any) are linked with broader changes in delay of gratification, (2) impacts of this intervention in more diverse samples, and (3) whether experimental manipulation of delay of gratification affects outcomes like achievement and weight, which have been linked to this skill in observational studies.

As an avid chess player, I found this dataset to be intriguing and unique as I mostly would find Chess and board games based datasets mostly without any sort of physical or real-life property to it as most of the chess or game datasets are entirely virtual. This dataset brings the real world in and would be a great challenge to make predictions using computer vision based networks. Thank you to Robo Flow for providing this dataset!

Andrew's Preflop Calls

This dataset contains (preflop call amount, hand) tuples for my No Limits Texas Hold'em Poker games. The data was collected over a period of one year, and includes over 10,000 hands.

The dataset can be used to study the preflop calling behavior of poker players. It can also be used to develop strategies for preflop calling.

Data

The dataset consists of two files:

• preflop_calls.csv: This file contains the (preflop call amount, hand) tuples.
• hands.csv: This file contains the full hand histories for the hands in the preflop_calls.csv file.

The preflop_calls.csv file has the following columns:

• hand: The hand that was played.
• preflop_call_amount: The amount of money that was called preflop.

The hands.csv file has the following columns:

• hand: The hand that was played.
• board: The board cards.
• players: The players in the hand.
• actions: The actions taken by the players in the hand.

Usage

The dataset can be used to study the preflop calling behavior of poker players. It can also be used to develop strategies for preflop calling.

To study the preflop calling behavior of poker players, you can use the preflop_calls.csv file to create a histogram of the preflop call amounts. You can also use the preflop_calls.csv file to create a scatterplot of the preflop call amounts against the hand strength.

To develop strategies for preflop calling, you can use the preflop_calls.csv file to identify the factors that are most predictive of preflop calling. You can then use these factors to develop a model that predicts preflop calling.

References

• No Limit Texas Hold'em Poker
• Preflop Calling

123