---
library_name: keras
license: mit
pipeline_tag: image-classification
datasets:
- ryanlinjui/darkchess-board
- ryanlinjui/darkchess-single
---

## Model description
This model serves as the **Eye of Darkchess Robot**, used to recognize the real-world Darkchess board.
> for more imformation, please visit: https://github.com/ryanlinjui/darkchess-robot

### Architecture
Based on [VGGNet](https://arxiv.org/pdf/1409.1556), with modifications to its architecture and parameters.

<details>
  <summary>Click me</summary>
  
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/64e8245d2c506d28d480fb46/W0Nk5n8NZRBRTKTlTZxV_.png)

</details>

## Intended uses & limitations

### Intended uses
- Darkchess Image Classification.
- Darkchess Board Image Recognition.

### Limitations
- Specifically for [Darkchess Robot](https://github.com/ryanlinjui/darkchess-robot).
- Non-directly used, preprocessing is necessary.

### How to use
TBA

## Training and evaluation data
TBA

## Training procedure
TBA

### Training hyperparameters

The following hyperparameters were used during training:

| Hyperparameters | Value |
| :-- | :-- |
| name | Adam |
| learning_rate | 9.999999747378752e-06 |
| decay | 0.0 |
| beta_1 | 0.8999999761581421 |
| beta_2 | 0.9990000128746033 |
| epsilon | 1e-08 |
| amsgrad | False |
| training_precision | float32 |