API Documentation

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

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

  • theano.function()[source]
  • Alias for function.function()
  • class theano.In[source]
  • Alias for function.In
  • theano.clone(output, replace=None, strict=True, share_inputs=True, copy_inputs=)[source]
  • Function that allows replacing subgraphs of a computational graph.
  • It returns a copy of the initial subgraph with the correspondingsubstitutions.

    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 originalgraph. If False, clone them. Note that cloned shared variables stilluse the same underlying storage, so they will always have the samevalue.
    • copy_inputs – Deprecated, use share_inputs.

    • theano.sparsegrad(_var)[source]
    • This function return a new variable whose gradient will bestored 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.