GU-CLASP / Typedflow

#+TITLE: TypedFlow

TypedFlow is a typed, higher-order frontend to [[http://www.tensorflow.org][TensorFlow]] and a high-level library for deep-learning.

The main design principles are:

• To make the parameters of layers explicit. This choice makes sharing of parameters explicit and allows to implement "layers" as pure functions.

• To provide as precise as possible types. Functions are explicit about the shapes and elements of the tensors that they manipulate (they are often polymorphic in shapes and elements though.)

• To let combinators be as transparent as possible. If a NN layers is a simple tensor transformation it will be exposed as such.

In this version, the interface to TensorFlow is done via python-code generation and a suitable runtime system.

** Documentation

The compiled documentation should be found on [[https://hackage.haskell.org/package/typedflow][hackage]].

** Examples

TypedFlow comes with two examples of neural networks:

• An adaptation of the [[examples/mnist][MNIST tensorflow tutorial]]
• A simple [[examples/seq2seq][sequence to sequence model]] which attempts to learn to translate pre-order into post-order.

To running the examples can be done like so:

#+BEGIN_SRC shell nix-env -iA nixpkgs.haskellPackages.styx nix-env -iA nixpkgs.cabal2nix styx configure cd examples/seq2seq make #+END_SRC