wiseodd/generative-models Alternatives

choose when you want a config-driven PyTorch GAN training/evaluation library rather than copy-paste snippets; GAN-only and no recent updates.

StudioGAN is a Pytorch library providing implementations of representative Generative Adversarial Networks (GANs) for conditional/unconditional image generation.

choose when you prefer Keras/TensorFlow high-level GAN examples over PyTorch/TensorFlow low-level scripts.

Keras implementations of Generative Adversarial Networks.

use when you want GAN code as book-companion notebooks rather than the multi-model seed.

Companion repository to GANs in Action: Deep learning with Generative Adversarial Networks

use when you want a broad GAN-only set of PyTorch examples and can accept stale code.

PyTorch implementations of Generative Adversarial Networks.

use when you want classic minimal TensorFlow examples of DCGAN/VAE/DRAW; expect outdated APIs.

A collection of generative methods implemented with TensorFlow (Deep Convolutional Generative Adversarial Networks (DCGAN), Variational Autoencoder (VAE) and DRAW: A Recurrent Neural Network For Image Generation).

use when you need TensorFlow Boltzmann-machine implementations without the seed's other models.

Boltzmann Machines in TensorFlow with examples

use when you need only the simplest vanilla GAN in PyTorch to start from.

A very simple generative adversarial network (GAN) in PyTorch

use when you need a simple standalone VAE in PyTorch and not the full collection.

A Variational Autoencoder (VAE) implemented in PyTorch

use when you want annotated notebook-style PyTorch examples of GAN/VAE variants only.

Annotated, understandable, and visually interpretable PyTorch implementations of: VAE, BIRVAE, NSGAN, MMGAN, WGAN, WGANGP, LSGAN, DRAGAN, BEGAN, RaGAN, InfoGAN, fGAN, FisherGAN

use when you need a standalone DiscoGAN implementation rather than a broader zoo.

PyTorch implementation of "Learning to Discover Cross-Domain Relations with Generative Adversarial Networks"

use when you need the canonical DCGAN TensorFlow implementation as a standalone example.

A tensorflow implementation of "Deep Convolutional Generative Adversarial Networks"

use when you need RBM/DBN models in C++ as a maintained library rather than Python snippets.

Fast Deep Learning Library (DLL) for C++ (ANNs, CNNs, RBMs, DBNs...)

use when you need a hyperspherical VAE implementation specifically.

Pytorch implementation of Hyperspherical Variational Auto-Encoders

use when you need a compact VQ-VAE implementation specifically.

Minimalist implementation of VQ-VAE in Pytorch

use when you want a TensorFlow-only collection of generative models, accepting older code.

Collection of generative models in Tensorflow

use when you want a beginner PyTorch VAE tutorial rather than a broad collection.

A simple tutorial of Variational AutoEncoders with Pytorch

use when you need a standalone TensorFlow implementation of BEGAN.

Tensorflow implementation of "BEGAN: Boundary Equilibrium Generative Adversarial Networks"

use when you want current, minimal PyTorch VAE/CVAE code for MNIST rather than the seed's broader set.

Variational Autoencoder and Conditional Variational Autoencoder on MNIST in PyTorch

use when you need a TensorFlow GAN for speech enhancement rather than general image generative models.

Speech Enhancement Generative Adversarial Network in TensorFlow

use when you need a minimal TensorFlow VAE example for MNIST.

Tensorflow implementation of variational auto-encoder for MNIST

use when you need a PyTorch VAE for text/sentences rather than image examples.

PyTorch Re-Implementation of "Generating Sentences from a Continuous Space" by Bowman et al 2015 https://arxiv.org/abs/1511.06349

use when you need a standalone VAE in TensorFlow/PyTorch with inverse autoregressive flow.

Variational autoencoder implemented in tensorflow and pytorch (including inverse autoregressive flow)

use when you want a deep VAE-only PyTorch collection rather than the seed's broad but shallow collection.

A Collection of Variational Autoencoders (VAE) in PyTorch.

use when you want a simple Keras VAE/CVAE example instead of separate PyTorch/TensorFlow code.

a simple vae and cvae from keras