Assert that actual.startswith(expected_start) is True. Args: actual: str expected_start: str msg: Optional message to report on

Creates a `RaggedTensor` with rows partitioned by `row_splits`. The returned `RaggedTensor` corresponds with the python list defined by: ```

Calculate the GPU launch config we should use for a kernel launch. This is assuming the kernel is quite simple and will largely be memory-limited. REQ

Return new `LinearOperator` acting like `op1 + op2`. Args: op1: `LinearOperator` op2: `LinearOperator`, with `shape` and `dtype` su

Returns the list of colocation groups of the op.

Return a slot named `name` created for `var` by the Optimizer. Some `Optimizer` subclasses use additional variables. For example `Momentum`

Precondition: !IsUninitialized(). It's very common to do foo.builder()->bar(). Without this precondition, if foo.builder() is null, the call to bar

Use cached_session instead.

Helper to create a WrappedTensor object.

Split strings using any of the supplied delimiters. For example: Split("a,b.c,d", ".,") would return {"a", "b", "c", "d"}.

The function to predict.

Returns a `Tensor` for the given key. A `str` key is used to access a base feature (not-transformed). When a `_FeatureColumn` is passed, the

Returns the next record (key, value pair) produced by the reader. The multiple reader instances are all configured to `read()` from the filen

Helper to make a zip file of a bunch of TensorFlow models. This does a cartestian product of the dictionary of test_parameters and calls make_gra

Enqueues a constant with the value of the given literal onto the computation.

Adds {name, value} pair to hyperparameters. Args: name: Name of the hyperparameter. value: Value of the hyperparameter. Can be one of

Removes the registered function corresponding to `token`.

Returns the resource manager associated w/ this device.

xla::int64 is not the same type as tensorflow::protobuf_int64 in open-source. Wrapper function that gives an int64 array slice view of a repeated int6

Create a test session with master target set to the testing cluster. Creates a test session that connects to the local testing cluster. The s

Removes the last layer in the model. Raises: TypeError: if there are no layers in the model.

Gets one element from this staging area. If the staging area is empty when this operation executes, it will block until there is an element t

The number of dimensions in this shape, or None if unknown.

Runs the core tests. Args: ds_fn: 0-argument function that returns a Dataset. num_outputs: Total number of outputs expected from this

Apply gradients to variables. This is the second part of `minimize()`. It returns an `Operation` that applies gradients. Args: gra

Returns a deep copy of node (excluding some fields, see copy_clean).

Get the name of Mkl op from original TensorFlow op We prefix 'Mkl' to the original op to get Mkl op.

Returns a list of integers or `None` for each dimension. Returns: A list of integers or `None` for each dimension. Raises: Value

Load the MetaGraphDef graph and restore variable values into the session. Args: sess: tf.compat.v1.Session to restore variable values.

Push 'v'. If the maximum number of elements was exceeded, drop the lowest element and return it in 'dropped' (if given). If the maximum is not exceed

Gets the computations in this module. Returns a view of HloComputation*s, so you can iterate over this in the natural way: for (HloComputation* c :

Creates an `Operation` in this graph. This is a low-level interface for creating an `Operation`. Most programs will not call this method dire