Gen Quick Start

This repository contains Jupyter notebooks that contain tutorials on specific features and/or applications of Gen. The notebooks walk you through programs that use Gen. Some notebooks also include exercises that expect you to write code or fill in written responses.

These notebooks assume some familiarity with the Julia programming language.

For reference on Gen see:

• The Gen documentation

• Documentation for GenPyTorch.jl and GenTF.jl

• Other packages in the Gen ecosystem

You can also find HTML rendered versions of the tutorial notebooks in this repository on the Gen website.

Getting started

To use the Jupyter notebooks in this repository, you will need to install Jupyter.

Gen is a package for the Julia language, so you will also need to download and install Julia. (Click the "help" link under your OS for specific installation instructions; we recommend following the instructions to "add Julia to PATH.")

Clone this repository, and inside the gen-quickstart directory, run

$ JULIA_PROJECT=. julia -e 'import Pkg; Pkg.instantiate()'
$ JULIA_PROJECT=. julia -e 'import Pkg; Pkg.build()'

to install Gen and its dependencies. You can then run

$ JULIA_PROJECT=. julia -e 'using IJulia; notebook(; dir=".")'

to start Jupyter. (It should also work to run jupyter notebook from your shell.) Navigate to the Tutorials notebook to get started!

Getting started on Google Colab

If you want to try gen immediately, click this goolge colab button!!

TensorFlow or PyTorch integration

Some of our tutorials use PyCall to interact with Python installations of PyTorch and TensorFlow. These tutorials are in their own directories with their own Julia Project.toml files, and require special setup.

To use them, make sure that python (version 3 only) is in your PATH and pointing to a Python environment that has PyTorch and TensorFlow installed. (This may in turn require installing other dependencies; please see PyTorch and TensorFlow websites for details. For more information on how PyCall and IJulia find the right versions of python and jupyter, please see the respective documentation of those packages.) Change into the tutorial directory of interest (e.g., tutorials/pytorch), and run:

$ export PYTHON=`which python`
$ JULIA_PROJECT=. julia -e 'import Pkg; Pkg.instantiate(); Pkg.build()'

When Julia builds PyCall, it should use the version of Python pointed to by your PYTHON environment variable.