aeppl.transforms.TransformedVariable#

class aeppl.transforms.TransformedVariable[source]#

A no-op that identifies a transform and its un-transformed input.

Methods

TransformedVariable.L_op(inputs, outputs, ...)

Construct a graph for the L-operator.

TransformedVariable.R_op(inputs, eval_points)

Construct a graph for the R-operator.

TransformedVariable.__init__()

TransformedVariable.add_tag_trace([user_line])

Add tag.trace to a node or variable.

TransformedVariable.connection_pattern(node)

TransformedVariable.do_constant_folding(...)

Determine whether or not constant folding should be performed for the given node.

TransformedVariable.get_params(node)

Try to get parameters for the Op when Op.params_type is set to a ParamsType.

TransformedVariable.grad(args, g_outs)

Construct a graph for the gradient with respect to each input variable.

TransformedVariable.infer_shape(fgraph, ...)

TransformedVariable.make_node(tran_value, value)

Construct an Apply node that represent the application of this operation to the given inputs.

TransformedVariable.make_py_thunk(node, ...)

Make a Python thunk.

TransformedVariable.make_thunk(node, ...[, impl])

Create a thunk.

TransformedVariable.perform(node, inputs, ...)

Calculate the function on the inputs and put the variables in the output storage.

TransformedVariable.prepare_node(node, ...)

Make any special modifications that the Op needs before doing Op.make_thunk.

Attributes

default_output

An int that specifies which output Op.__call__ should return.

destroy_map

A dict that maps output indices to the input indices upon which they operate in-place.

itypes

otypes

params_type

view_map

A dict that maps output indices to the input indices of which they are a view.