This is a repository containing code for implementing dropout in neural networks on multiple datasets. Dropout is a regularization technique that can prevent overfitting in neural networks by randomly dropping out some neurons during training.
The following datasets are included in this repository:
- MNIST
- CIFAR-10
- Boston Housing Price Regression
- Fashion-MNIST
- Street View House Numbers
- Reuters RCV1
To run the code in this repository, you will need the following dependencies:
- Python 3
- TensorFlow
- Keras
- Numpy
- Matplotlib
To use this code, simply clone the repository and run the desired script for the dataset you wish to use. For example, to run the MNIST script, use the following command:
python mnist_dropout.py
This will train a neural network on the MNIST dataset with dropout regularization.
The results of each experiment can be found in the results folder. Each result file includes the accuracy and loss of the neural network during training and testing.