private:
  TensorTape tensor_tape_;
  OpTape<BackwardFunction, TapeTensor> op_tape_;
  int64_t next_op_id_{0};

  // Map from tensor id to number of remaining usages (i.e. how many entries in
  // the tape refer to it); to aid in tape garbage collection.
  std::unordered_map<int64_t, int64_t> tensor_usage_;

  // If false, all activations are deleted in the first call to ComputeGradient.
  // Else, only when this is destructed.

       * Refer [windows setup guide](https://www.tensorflow.org/install/gpu#windows_setup).
     * If you have above configuration and using _**Ubuntu/Linux**_ platform -
       * Try adding the CUDA, CUPTI, and cuDNN installation directories to the $LD_LIBRARY_PATH environment variable.
       * Refer [linux setup guide](https://www.tensorflow.org/install/gpu#linux_setup).

        base: ${{ github.event.inputs.release_branch }}
        branch: ${{ github.event.inputs.release_branch }}-${{ steps.cherrypick.outputs.SHORTSHA }}
        reviewers: learning-to-play
        body: |

//       `TP_OptimizerRegistrationParams*` as input with `struct_size`
//       populated by core. The plugin is responsible for setting
//       `struct_size` as well, along with all other fields.
//     * Refer to "TensorFlow Versioning Strategy" section at
//       https://github.com/tensorflow/community/pull/257/files.
//     * Note that the API is still under active development and doesn't have
//       versioning guarantees yet.

may exist in your changes.

#### Running unit tests

There are two ways to run TensorFlow unit tests.

1.  Using tools and libraries installed directly on your system.

    Refer to the
    [CPU-only developer Dockerfile](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/tools/dockerfiles/dockerfiles/devel-cpu.Dockerfile)
    and

        (neural) layers.
    *   Add `tf.contrib.bayesflow.halton_sequence`.
    *   Add `tf.contrib.data.make_saveable_from_iterator.`
    *   Add `tf.contrib.data.shuffle_and_repeat`.
    *   Add new custom transformation: `tf.contrib.data.scan()`.
    *   `tf.contrib.distributions.bijectors`:
    *   Add `tf.contrib.distributions.bijectors.MaskedAutoregressiveFlow`.

  // Returns the operation's device name.
  //
  // The value returned may be different from the one set by SetDeviceName, but
  // it will be compatible with it: the name will be updated by device placement
  // logic to refer to the specific device chosen.
  //
  // Example: If one calls `op->SetDeviceName("/device:GPU")`, the value
  // returned by DeviceName should be "/device:GPU:*" until a particular GPU is

Packages are imported one by one in alphabetical order during runtime.

The test doesn't identify package's order-dependent issues; for instance,
importing "tf.foo" followed by "tf.bar" won't reveal that "tf.bar" depends on
"tf.foo" being imported first.

The `_api/v2/api_packages.txt` file is generated during the TensorFlow API v2
init files creation process and is subsequently stored in the wheel file after

# limitations under the License.
# ==============================================================================
"""
Top-level module of TensorFlow. By convention, we refer to this module as
`tf` instead of `tensorflow`, following the common practice of importing
TensorFlow via the command `import tensorflow as tf`.

The primary function of this module is to import all of the public TensorFlow

// A handle to a tensor on a device.
//
// Like a TF_Tensor, a TFE_TensorHandle refers to a tensor with a value, shape,
// type etc. Unlike a TF_Tensor, a TFE_TensorHandle may refer to such tensors
// placed in the memory of different devices or remote address spaces.
typedef struct TFE_TensorHandle TFE_TensorHandle;

TF_CAPI_EXPORT extern TFE_TensorHandle* TFE_NewTensorHandle(const TF_Tensor* t,