All Projects → HyeongminLEE → Tensorflow_DCGAN

HyeongminLEE / Tensorflow_DCGAN

Licence: other
Study Friendly Implementation of DCGAN in Tensorflow

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Tensorflow DCGAN

Gan Tutorial
Simple Implementation of many GAN models with PyTorch.
Stars: ✭ 227 (+931.82%)
Mutual labels:  gan, dcgan, celeba
Pytorch Mnist Celeba Gan Dcgan
Pytorch implementation of Generative Adversarial Networks (GAN) and Deep Convolutional Generative Adversarial Networks (DCGAN) for MNIST and CelebA datasets
Stars: ✭ 363 (+1550%)
Mutual labels:  gan, dcgan, celeba
DCGAN-CelebA-PyTorch-CPP
DCGAN Implementation using PyTorch in both C++ and Python
Stars: ✭ 14 (-36.36%)
Mutual labels:  gan, dcgan, celeba
Tf.gans Comparison
Implementations of (theoretical) generative adversarial networks and comparison without cherry-picking
Stars: ✭ 477 (+2068.18%)
Mutual labels:  gan, dcgan, celeba
Pycadl
Python package with source code from the course "Creative Applications of Deep Learning w/ TensorFlow"
Stars: ✭ 356 (+1518.18%)
Mutual labels:  gan, dcgan, celeba
Tf Exercise Gan
Tensorflow implementation of different GANs and their comparisions
Stars: ✭ 110 (+400%)
Mutual labels:  gan, dcgan, celeba
Face generator
DCGAN face generator 🧑.
Stars: ✭ 146 (+563.64%)
Mutual labels:  gan, dcgan
Gan
Resources and Implementations of Generative Adversarial Nets: GAN, DCGAN, WGAN, CGAN, InfoGAN
Stars: ✭ 2,127 (+9568.18%)
Mutual labels:  gan, dcgan
Image generator
DCGAN image generator 🖼️.
Stars: ✭ 173 (+686.36%)
Mutual labels:  gan, dcgan
Anogan Tf
Unofficial Tensorflow Implementation of AnoGAN (Anomaly GAN)
Stars: ✭ 218 (+890.91%)
Mutual labels:  gan, dcgan
Msg Gan V1
MSG-GAN: Multi-Scale Gradients GAN (Architecture inspired from ProGAN but doesn't use layer-wise growing)
Stars: ✭ 116 (+427.27%)
Mutual labels:  gan, celeba
Catdcgan
A DCGAN that generate Cat pictures 🐱‍💻
Stars: ✭ 177 (+704.55%)
Mutual labels:  gan, dcgan
DLSS
Deep Learning Super Sampling with Deep Convolutional Generative Adversarial Networks.
Stars: ✭ 88 (+300%)
Mutual labels:  gan, dcgan
Semantic image inpainting
Semantic Image Inpainting
Stars: ✭ 140 (+536.36%)
Mutual labels:  gan, dcgan
Generative adversarial networks 101
Keras implementations of Generative Adversarial Networks. GANs, DCGAN, CGAN, CCGAN, WGAN and LSGAN models with MNIST and CIFAR-10 datasets.
Stars: ✭ 138 (+527.27%)
Mutual labels:  gan, dcgan
Tensorflow Mnist Gan Dcgan
Tensorflow implementation of Generative Adversarial Networks (GAN) and Deep Convolutional Generative Adversarial Netwokrs for MNIST dataset.
Stars: ✭ 163 (+640.91%)
Mutual labels:  gan, dcgan
Deep Learning With Python
Example projects I completed to understand Deep Learning techniques with Tensorflow. Please note that I do no longer maintain this repository.
Stars: ✭ 134 (+509.09%)
Mutual labels:  gan, dcgan
GANs-Keras
GANs Implementations in Keras
Stars: ✭ 24 (+9.09%)
Mutual labels:  gan, dcgan
cDCGAN
PyTorch implementation of Conditional Deep Convolutional Generative Adversarial Networks (cDCGAN)
Stars: ✭ 49 (+122.73%)
Mutual labels:  dcgan, celeba
catgan pytorch
Unsupervised and Semi-supervised Learning with Categorical Generative Adversarial Networks
Stars: ✭ 50 (+127.27%)
Mutual labels:  gan, dcgan

DCGAN in Tensorflow

Basic Implementation (Study friendly) of DCGAN in Tensorflow

[Paper | Post(in Korean) | Pytorch Version]

1. Environments

  • Windows 10
  • Python 3.5.3 (Anaconda)
  • Tensorflow 1.4.0
  • Numpy 1.13.1
  • lmdb (pip install lmdb): for LSUN Dataset
  • cv2 (conda install -c conda-forge opencv): for LSUN Dataset

2. Networks and Parameters

2.1 Hyper-Parameters

  • Image Size = 64x64 (Both in CelebA and LSUN-Bedroom)
  • Batch Size = 128 (~32 is OK)
  • Learning Rate = 0.0002
  • Adam_beta1 = 0.5
  • z_dim = 100
  • Epoch = 5 in CelebA is Enough, 1 in LSUN is Enough. Sometimes it can be diverge.

2.2 Generator Networks (network.py)

2.3 Discriminator Networks (network.py)

3. Run (Train)

You can modify hyper-parameter. Look at the parsing part of the code.

3. 1 CelebA DB (Cropped Face, 156253 Samples)

  • Database Setting: link

  • Train & Test

python train.py --filelist <filelist_name> --out_dir <output_directory>
  • Test results will be saved in 'output_directory'

3. 2 LSUN-Bedroom DB (3033042 Samples)

  • Database Setting: link

  • Train & Test

python train.py --filelist <filelist_name> --out_dir <output_directory>
  • Test results will be saved in 'output_directory'

4. Results

DCGAN with CelebA (6 Epochs)

DCGAN with LSUN (1 Epochs)

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].