Library Documentation

This documentation covers Theano module-wise. This is suited to finding the Types and Ops that you can use to build and compile expression graphs.

• tensor – Types and Ops for Symbolic numpy
• gradient – Symbolic Differentiation
• config – Theano Configuration
• printing – Graph Printing and Symbolic Print Statement
• compile – Transforming Expression Graphs to Functions
• sparse – Symbolic Sparse Matrices [doc TODO]
• scalar – Symbolic Scalar Types, Ops [doc TODO]
• gof – Theano Internals [doc TODO]
• scan – Looping in Theano

There are also some top-level imports that you might find more convenient:

Platforms: Unix, Windows

theano.function(...)

Alias for function.function()

theano.shared(...)

Alias for shared.shared()

class theano.Param

Alias for function.Param

theano.dot(x, y)

Works like tensor.dot() for both sparse and dense matrix products

theano.__src_version__()

Returns the mercurial (hg) identifier of the current tip.

If there are uncommitted modifications, this function raises an Exception. If the theano package is not under mercurial control, then again this function raises an Exception.