Loads all layer weights, either from a TensorFlow or an HDF5 file.

Start threads for `QueueRunners`. Note that the queue runners collected in the graph key `QUEUE_RUNNERS` are already started automatically wh

Overload to call Or with 3 or more operands. As with `And`, we need the following complicated overload set to handle the default arg in the `Or` over

Reports error message when the reporter is set.

Fully connected MLP parameterized via `real_nvp_template`.

Returns the context in which computations are called at this call site.

Writes `MetaGraphDef` to save_path/filename. Args: filename: Optional meta_graph filename including the path. collection_list: List o

Retrieves updates relevant to a specific set of inputs. Arguments: inputs: Input tensor or list/tuple of input tensors. Returns:

Saves all layer weights. Either saves in HDF5 or in TensorFlow format based on the `save_format` argument. When saving in HDF5 format, t

Compute the number of elements in this table. Args: name: A name for the operation (optional). Returns: A scalar tensor containi

Err converts the status to a Go error and returns nil if the status is OK.

Initialize handle and devices if not already done so.

Log probability density/mass function. Args: value: `float` or `double` `Tensor`. name: Python `str` prepended to names of ops create

Private method used to set an attribute in the node_def.

Add a sequence of inputs to the function invocation. Args: *args: List of inputs to be converted (should be Tf.Tensor). **kwargs: Thi

Get the captured writes as a single string.

Creates a scalar String tensor using the default, UTF-8 encoding. @param data The string to put into the new scalar tensor.

Reshape the input to the target shape. If called several times, the axes named in existing_axis_names must be identical. Args: lab

Returns the value of the fuzz factor used in numeric expressions. Returns: A float. Example: ```python keras.backend.epsilon() >>>1e-0

Returns an `Optional` that wraps the given value. Args: value: A value to wrap. The value must be convertible to `Tensor` or `Compo

Methods for accessing the layout field.

Returns the number of components in a queue-element tuple.

The size of the i-th dimension. @return The size of the requested dimension or -1 if it is unknown.

Merges two summaries through an algorithm that's derived from MergeSort for summary entries while guaranteeing that the max approximation error of the

Matches the first output of a node that matches `node`.

SubScope returns a new Scope which will cause all operations added to the graph to be namespaced with 'namespace'. If namespace collides with an exis

DO NOT USE THIS FUNCTION FOR NEW FUNCTIONALITY BEYOND IMPLEMENTING BROADCASTING. Same as Offset(), except takes as NdArrayDesc<N> instead of Dims<N>.

Gets the transformed feature column.

Get whether this HloValue is live out of the module.

Runs model inference if the model takes only one input, and provides only one output. <p>Warning: The API runs much faster if {@link ByteBuffer} is u