https://github.com/hcnoh/dcgan-tensorflow2

TensorFlow2 Implementation for DCGAN
https://github.com/hcnoh/dcgan-tensorflow2

deep-learning neural-networks tensorflow2 tensorflow2-experiments tensorflow2-models

Last synced: about 2 months ago
JSON representation

TensorFlow2 Implementation for DCGAN

• Host: GitHub
• URL: https://github.com/hcnoh/dcgan-tensorflow2
• Owner: hcnoh
• Created: 2019-10-18T05:18:03.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2022-11-22T04:36:12.000Z (about 2 years ago)
• Last Synced: 2023-03-08T23:13:27.613Z (almost 2 years ago)
• Topics: deep-learning, neural-networks, tensorflow2, tensorflow2-experiments, tensorflow2-models
• Language: Jupyter Notebook
• Homepage:
• Size: 19 MB
• Stars: 0
• Watchers: 2
• Forks: 1
• Open Issues: 3
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# DCGAN Implementation for Generating MNIST Images in TensorFlow2

This repository is for the TensorFlow2 implementation for DCGAN. This repository provides the training module and Jupyter notebook for testing a generation of the trained models. MNIST dataset was used for this repository.

![](/assets/img/README/README_2019-10-26-18-44-22.png)

## Install Dependencies

1. Install Python 3.5.2.

2. Install TensorFlow ver 2.0.0. If you can use a GPU machine, install the GPU version of TensorFlow, or just install the CPU version of it.

3. Install Python packages(Requirements). You can install them simply by using following bash command.

    ```bash

    $ pip install -r requirements

    ```

    You can use `Virtualenv`, so that the packages for requirements can be managed easily. If your machine have `Virtualenv` package, the following bash command would be useful.

    ```bash

    $ virtualenv dcgan-tf2-venv

    $ source ./dcgan-tf2-venv/bin/activate

    $ pip install -r requirements.txt

    ```

## Training

*Note: TensorFlow provides dataset modules for some well known datasets such as MNIST, CIFAR-10 etc. In this repository, the only usage for TensorFlow MNIST dataset module was implemented yet. Usages for other datasets will be implemented too.*

1. **Modify the path for dataset in `config.py`.**

2. **Modify the path for directory for saving model checkpoint.**

3. **Execute training process by `train.py`.**

## Checking Results and Testing Generation

The Jupyter notebook for checking results and testing the image generation is provided. Please check `result_plot.ipynb`.

## Results

1. **Ploting the Generator and Discriminator Losses**

    ![](/assets/img/README/README_2019-10-26-18-50-34.png)

2. **Image Generation Results**

    ![](/assets/img/result_plot/image_generation_result_changes.gif)

## References

- GAN: [Generative Adversarial Nets](http://papers.nips.cc/paper/5423-generative-adversarial-nets)

- DCGAN : [Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://arxiv.org/abs/1511.06434)

## Author

Hyungcheol Non / [About Me](https://hcnoh.github.io/about)