https://github.com/roma-glushko/rock-paper-scissors
🎲 Application of Computer Vision to the classic game Rock-Paper-Scissors
https://github.com/roma-glushko/rock-paper-scissors
computer-vision machine-learning mobilenetv2 rock-paper-scissors tensorflow
Last synced: about 2 months ago
JSON representation
🎲 Application of Computer Vision to the classic game Rock-Paper-Scissors
- Host: GitHub
- URL: https://github.com/roma-glushko/rock-paper-scissors
- Owner: roma-glushko
- License: mit
- Created: 2021-01-30T21:26:42.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2021-09-29T18:58:24.000Z (about 4 years ago)
- Last Synced: 2025-02-28T08:33:58.799Z (7 months ago)
- Topics: computer-vision, machine-learning, mobilenetv2, rock-paper-scissors, tensorflow
- Language: Jupyter Notebook
- Homepage:
- Size: 7.88 MB
- Stars: 6
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
- License: LICENSE
Awesome Lists containing this project
README
# Rock-Paper-Scissors Recognition
Application of Computer Vision to the classic game Rock-Paper-Scissors.
This repository includes dataset analysis and modeling part of the task.
The model will be deployed as a JavaScript application on my website.
Dataset: https://www.kaggle.com/frtgnn/rock-paper-scissor
## Installation
```bash
poetry install
cd data
kaggle datasets download --unzip frtgnn/rock-paper-scissor
```
## Modeling
The best scores I was able to achieve with the following configs:
- Fully Freezed MobileNetV2 + RMSProp + L2Regularization(0.01) - ?
- Fully Freezed MobileNetV2 + AdamW(weight_decay: 0.01) - ?
## References
- https://keras.io/guides/transfer_learning/
- https://www.tensorflow.org/tutorials/images/transfer_learning
- https://ruder.io/transfer-learning/
- http://www.laurencemoroney.com/rock-paper-scissors-dataset/