https://github.com/praatibhsurana/digitgan

Implementing a Generative Adversarial Network using Keras.
https://github.com/praatibhsurana/digitgan

deep-learning gan keras neural-networks python

Last synced: about 2 months ago
JSON representation

Implementing a Generative Adversarial Network using Keras.

• Host: GitHub
• URL: https://github.com/praatibhsurana/digitgan
• Owner: praatibhsurana
• Created: 2020-08-25T18:14:36.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-09-01T13:57:34.000Z (about 5 years ago)
• Last Synced: 2025-03-31T01:47:58.752Z (7 months ago)
• Topics: deep-learning, gan, keras, neural-networks, python
• Language: Python
• Homepage:
• Size: 507 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# DigitGAN

Keras implementation of a simple General Adversarial Network (GAN) to generate digits.

The mnist dataset was used.

## Setup

**Clone the repo and navigate to it**

```bash

git clone https://github.com/praatibhsurana/DigitGAN.git

cd DigitGAN-master

```

**To run the model**

- Download [pip](https://pip.pypa.io/en/stable/installing/)

- Download [python 3](https://www.python.org/downloads/)

- Create and activate your virtual environment 

- From the root directory run:

```bash

    pip install -r requirements.txt

```

- You should be good to go now! Now, from within the virtual environment you created, run:

```bash

    python DigitGAN.py

```

The quality of generated images can be improved by running for a higher number of epochs.

## Results obtained from various epochs-

### Epoch 1

![1](https://github.com/praatibhsurana/DigitGAN/blob/master/Generator_Checkpoints/gan_generated_image_epoch_1.png?raw=true)

### Epoch 10

![2](https://github.com/praatibhsurana/DigitGAN/blob/master/Generator_Checkpoints/gan_generated_image_epoch_10.png?raw=true)

### Epoch 20

![3](https://github.com/praatibhsurana/DigitGAN/blob/master/Generator_Checkpoints/gan_generated_image_epoch_20.png?raw=true)

### Epoch 30

![4](https://github.com/praatibhsurana/DigitGAN/blob/master/Generator_Checkpoints/gan_generated_image_epoch_30.png?raw=true)

### Epoch 40

![5](https://github.com/praatibhsurana/DigitGAN/blob/master/Generator_Checkpoints/gan_generated_image_epoch_40.png?raw=true)

### Epoch 50

![6](https://github.com/praatibhsurana/DigitGAN/blob/master/Generator_Checkpoints/gan_generated_image_epoch_50.png?raw=true)