App I made for The Odin Project.
- Basic of html,css and javascript
- basic working of websocket
- implementing neuralnet from scratch
- gradient descent algorithm for neuralnet
- sigmoid function
- forward and backward propagation of neuralnet
- Add Regularization for neuralnet for better performance(prevent overfitting)
- Improve the training data and optimized the neuralnet for better performance (tuning hyperparamters)
- Websocket implemented more than two players can play tic tac toe over the network.
- Neuralnet is integrated with websocket on python
- User can play with neuralnet and they can also optimize its performance by fine tuning the hyperparameter
npm i
npm run dev
cd neuralnet_server
pip install -r requirements.txt
python3 neuralnet_server/server.py
Took the code reference for neuralnet from here
To train the neuralnet
python3 -m neuralnet_server.model.train.py
To pretrain the neuralnet and generate training dataset(need to uncomment few things you can check on code)
python3 -m neuranet_server.model.play_train.pyMIT