HyperGAN
  • About
  • Getting started
  • CLI guide
  • Configurations
    • Configurable Parameters
  • Showcase
    • AI Explorer for Android
    • Youtube, Twitter, Discord +
  • Examples
    • 2D
    • Text
    • Classification
    • Colorizer
    • Next Frame (video)
  • Tutorials
    • Training a GAN
    • Pygame inference
    • Creating an image dataset
    • Searching for hyperparameters
  • Components
    • GAN
      • Aligned GAN
      • Aligned Interpolated GAN
      • Standard GAN
    • Generator
      • Configurable Generator
      • DCGAN Generator
      • Resizable Generator
    • Discriminator
      • DCGAN Discriminator
      • Configurable Discriminator
    • Layers
      • add
      • cat
      • channel_attention
      • ez_norm
      • layer
      • mul
      • multi_head_attention
      • operation
      • pixel_shuffle
      • residual
      • resizable_stack
      • segment_softmax
      • upsample
    • Loss
      • ALI Loss
      • F Divergence Loss
      • Least Squares Loss
      • Logistic Loss
      • QP Loss
      • RAGAN Loss
      • Realness Loss
      • Softmax Loss
      • Standard Loss
      • Wasserstein Loss
    • Latent
      • Uniform Distribution
    • Trainer
      • Alternating Trainer
      • Simultaneous Trainer
      • Balanced Trainer
      • Accumulate Gradient Trainer
    • Optimizer
    • Train Hook
      • Adversarial Norm
      • Weight Constraint
      • Stabilizing Training
      • JARE
      • Learning Rate Dropout
      • Gradient Penalty
      • Rolling Memory
    • Other GAN implementations
Powered by GitBook
On this page
  • Acquiring a dataset
  • Getting started
  • Testing install
  • Training hypergan
  • Sampling your model
  • Building your model
  • Deploying your model
  • TODO

Was this helpful?

  1. Tutorials

Training a GAN

Thanks for trying out hypergan. This tutorial will walk you through creating your first model, exploring it, and building/deploying the generator.

Acquiring a dataset

You can acquire a dataset by having a folder full of loose images.

Getting started

First install hypergan.

Testing install

To test hypergan is installed correctly, run hypergan new . -l

You can view video card usage with nvidia-smi

Training hypergan

You can run:

  hypergan train /path/to/data -c myconfig  -s 256x256x3 -format png --sampler static_batch --sample_every 10

Once hypergan is working, you can sample your

Sampling your model

You've created an infinite manifold. Great! With any luck it's diverse. You can see how diverse your model is by using:

  hypergan sample /path/to/data -c myconfig -s 256x256x3 -format png --sampler batch_walk

Building your model

  hypergan build /path/to/data -c myconfig -s 256x256x3 -format png --sampler batch --sample_every 10

Deploying your model

You've made a great model you'd like to share. Awesome! You can deploy it to users through a few methods.

TODO

PreviousNext Frame (video)NextPygame inference

Last updated 4 years ago

Was this helpful?