This project streamlines the construction, training, and usage of deep generative models such as variational autoencoders, generative adversarial networks and autoregressive models for Bayesian image processing tasks. dgm-pipeline
is built around TensorFlow 2 and also makes use of tensorflow-probability
and tensorflow-datasets
.
I'm also using it to familiarize myself with software engineering techniques and tools such as test-driven development and continuous integration. This is definitely an instructional project. Feel free to contribute with a pull request!
The best way to install this repository is to clone it. There is currently no package available on PyPi.