Core operations

Types

The computation graph as a whole is stored in the Graph type. Individual nodes in the graph are referred to as operations, stored in the Operation type. Operations can produce one or more outputs, represented as the Tensor type.

# TensorFlow.Tensor — Type.

Represents the output of an operation in the computation graph

# TensorFlow.Operation — Type.

An operation in the computation graph.

# TensorFlow.Session — Type.

A TensorFlow session.

# TensorFlow.Graph — Type.

A TensorFlow computation graph

Functions

# Base.eltype — Function.

eltype(type)

Determine the type of the elements generated by iterating a collection of the given type. For associative collection types, this will be a Pair{KeyType,ValType}. The definition eltype(x) = eltype(typeof(x)) is provided for convenience so that instances can be passed instead of types. However the form that accepts a type argument should be defined for new types.

# TensorFlow.node_name — Function.

node_name(node::AbstractOperation)

Returns the name of a node in the computation graph.

# TensorFlow.get_collection — Function.

Returns a collection attached to the graph g named name

Returns a collection from the default graph

# TensorFlow.get_def_graph — Function.

Returns the default computation graph, an object of type Graph.

# Base.run — Function.

run(command)

Run a command object, constructed with backticks. Throws an error if anything goes wrong, including the process exiting with a non-zero status.

# TensorFlow.get_node_by_name — Function.

Returns an operation by searching for its name in the given graph.

# TensorFlow.ShapeInference.get_shape — Function.

Runs shape inference to return the shape of the tensor produced by the given operation.

Note this runs statically. Use the shape operation to dynamically get the shape of an operation.

# TensorFlow.get_def — Function.

Returns the definition of the given operation or graph