---
license: lgpl-3.0
language:
- en
pipeline_tag: image-feature-extraction
---
# Model Card for BoardCNN

BoardCNN implements a Convolutional Neural Network (CNN) to recognize the position from images of chess boards.

The model expects a board image as input and returns the expected positions of the pieces on the board.

## Model Details

Custom CNN architecture was implemented via pytorch 



**Developed by:** Igor Alexey <br>
**Model type:** Safetensors <br>
**License:** GNU GPL v3 <br>

### Model Sources 

- **Repository:** [More Information Needed]
- **Demo:** [More Information Needed]

## Uses

The model can be used to make predictions on new chess board images. The output is a 8x8 grid of chess piece symbols, representing the predicted position of pieces on the board.


### Out-of-Scope Use

The pre-trained models are not made for scanning 3D boards, although it's likely the architecture should scale well for this task with a proper training set.

## Limitations

Might not always give 100% correct output, especially on varying piece sets and board themes.


## Getting started

Use the code below to get started with the model.

[More Information Needed]

## Training Details

### Training Data

The models are trained on 5k gnerated images of valid random board positions with reasonable piece sets from lichess.

### Training Procedure

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->


#### Training Hyperparameters

- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->

#### Speeds, Sizes, Times [optional]

<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->

[More Information Needed]

## Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

### Testing Data, Factors & Metrics

#### Testing Data

<!-- This should link to a Dataset Card if possible. -->

[More Information Needed]

#### Factors

<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->

[More Information Needed]

#### Metrics

<!-- These are the evaluation metrics being used, ideally with a description of why. -->

[More Information Needed]

### Results

[More Information Needed]

#### Summary