API 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.

• compile – Transforming Expression Graphs to Functions
• config – Theano Configuration
• d3viz – d3viz: Interactive visualization of Theano compute graphs
• gof – Theano Internals [doc TODO]
• gpuarray – The (new) GPU backend
• gradient – Symbolic Differentiation
• misc.pkl_utils - Tools for serialization.
• printing – Graph Printing and Symbolic Print Statement
• sandbox – Experimental Code
• scalar – Symbolic Scalar Types, Ops [doc TODO]
• scan – Looping in Theano
• sparse – Symbolic Sparse Matrices
• sparse – Sparse Op
• sparse.sandbox – Sparse Op Sandbox
• tensor – Types and Ops for Symbolic numpy
• typed_list – Typed List
• tests – Tests

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

theano.function(...)

Alias for theano.compile.function.function()

theano.function_dump(...)

Alias for theano.compile.function.function_dump()

theano.shared(...)

Alias for theano.compile.sharedvalue.shared()

class theano.In

Alias for function.In

theano.dot(x, y)

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

theano.clone(output, replace=None, strict=True, share_inputs=True, copy_inputs=<object object>)

Function that allows replacing subgraphs of a computational graph.

It returns a copy of the initial subgraph with the corresponding substitutions.

Parameters

• output (Theano Variables (or Theano expressions)) – Theano expression that represents the computational graph.

• replace (dict) – Dictionary describing which subgraphs should be replaced by what.

• share_inputs (bool) – If True, use the same inputs (and shared variables) as the original graph. If False, clone them. Note that cloned shared variables still use the same underlying storage, so they will always have the same value.

• copy_inputs – Deprecated, use share_inputs.

theano.sparse_grad(var)

This function return a new variable whose gradient will be stored in a sparse format instead of dense.

Currently only variable created by AdvancedSubtensor1 is supported. i.e. a_tensor_var[an_int_vector].

New in version 0.6rc4.