We still have that one convenience accessor to avoid the awkward double bracket issue: shape.dims()[i].

Computes the gradient using operations recorded in context of this tape. Args: target: a list or nested structure of Tensors or Variables t

Returns the next record (key, value) pair produced by a reader. Will dequeue a work unit from queue if necessary (e.g. when the Reader needs

The graph-level random seed of this graph.

Methods for accessing the dimensions array.

Adds a layer instance on top of the layer stack. Arguments: layer: layer instance. Raises: TypeError: If `layer` is not a la

Creates an instruction pattern that will capture the matched instruction in the argument.

Returns `Tensor`'s shape partitioned into `sample`, `batch`, `event`. Args: x: `Tensor`. name: Python `str`. The name to give this op

Use cached_session instead.

Enter a context where the lock is with group `group_id`. Args: group_id: The group for which to acquire and release the lock. Returns:

Outputs random values from a uniform distribution. The generated values follow a uniform distribution in the range `[minval, maxval)`. The lo

The function to predict.

Returns a callable that returns the next element of the dataset. Example use: ```python # In both graph and eager modes dataset = ...

Convenience method for making a new DeviceSpec by overriding fields. For instance: ``` my_spec = DeviceSpec=(job="my_job", device="CPU")

Build the {@link Operation}. <p>The following action will also be performed depending on the current execution environment. <ul> <li>In eager mode

Returns a nested structure of `tf.Tensor`s representing the next element. In graph mode, you should typically call this method *once* and use its

Returns the number of regular fanins.

A transformation that models performance. Returns: A `Dataset` transformation function, which can be passed to `tf.data.Dataset.apply`.

Returns a mutable pointer to the `NodeItem` with the given `id` if it exists in the graph, or `nullptr` if it does not.

Returns a context manager that specifies an op to colocate with. Note: this function is not for public use, only for internal libraries. For

r'''support standalone evaluation.

Return selected values in the TensorArray as a packed `Tensor`. All of selected values must have been written and their shapes must all match

Adds a child piece to this piece's children.

Returns cuDNN handle. To be passed directly to cuDNN APIs, don't keep a copy.

Mean.

The `TensorShape` of this variable. Returns: A `TensorShape`.

Asserts that a dataset produces the expected output / error. Args: dataset: A dataset to check for the expected output / error. expec

Attr returns the value of an attribute on op. It returns an error if the attribute does not exist.

Stores `value` in the collection with the given `name`. Note that collections are not sets, so it is possible to add a value to a collection

Gets the instructions in this computation. The returned type is a range of HloInstruction*s, so you can iterate over it using a range-based for loop

Returns the (mutable) result shape of this instruction.

Sorts a tensor. Usage: ```python import tensorflow as tf a = [1, 10, 26.9, 2.8, 166.32, 62.3] b = tf.sort(a,axis=-1,direction='ASCENDING

Add a function definition to the context. Once added, the function (identified by its name) can be executed like any other operation. Ar

Return a pointer to the data pointers of all tensors in the list. For example: float* const* f = v.data(); f[0][1] is the second element of the first

Applies a transformation function to this dataset. `apply` enables chaining of custom `Dataset` transformations, which are represented as fun

Returns true if `this` represents an actual value.

Assign `value` to all replicas. Outside of the tpu.rewrite context, assign explicitly to all replicas. Inside of the tpu.rewrite context, ass

Const adds an operation to graph that produces value as output.

Saves variables. This method runs the ops added by the constructor for saving variables. It requires a session in which the graph was launche