CompareTensors(out1, expected_out1);
  }

  void CompareTensors(TF_Tensor* a, TF_Tensor* b) {
    float* a_data = static_cast<float*>(TF_TensorData(a));
    float* b_data = static_cast<float*>(TF_TensorData(b));
    EXPECT_EQ(*a_data, *b_data);
  }

  void AddGradients(bool grad_inputs_provided, const char* prefix,
                    TF_Output* inputs, int ninputs, TF_Output* outputs,

// Return a tensor handle with given type, values and dimensions.
template <class T, TF_DataType datatype>
Status TestTensorHandleWithDims(AbstractContext* ctx, const T* data,
                                const int64_t* dims, int num_dims,
                                AbstractTensorHandle** tensor) {
  std::unique_ptr<TF_Status, decltype(&TF_DeleteStatus)> status(

    python tf_doctest.py --file=<file_path>
    ```

    This will run it using your installed version of TensorFlow. To be sure
    you're running the same code that you're testing:

    *   Use an up to date [tf-nightly](https://pypi.org/project/tf-nightly/)
        `pip install -U tf-nightly`
    *   Rebase your pull request onto a recent pull from
        [TensorFlow's](https://github.com/tensorflow/tensorflow) master branch.

     */

    AbstractTensorHandle* upstream_grad = grad_outputs[0];
    DCHECK(upstream_grad);

    // Recover data format from forward pass for gradient.
    std::string data_format;
    TF_RETURN_IF_ERROR(forward_attrs_.Get("data_format", &data_format));

    // Grad for A
    grad_inputs[0] = upstream_grad;
    grad_inputs[0]->Ref();

    // Grad for bias

      auto* out_shape_and_type = handle_data.add_shape_and_type();
      ic->ShapeHandleToProto(p.shape, out_shape_and_type->mutable_shape());
      out_shape_and_type->set_dtype(p.dtype);
      *out_shape_and_type->mutable_type() = p.type;
    }
  }
  string str_data;
  handle_data.SerializeToString(&str_data);

  TF_Buffer* result = TF_NewBufferFromString(str_data.c_str(), str_data.size());
  return result;
}

tblib==2.0.0 \
    --hash=sha256:9100bfa016b047d5b980d66e7efed952fbd20bd85b56110aaf473cb97d18709a \
    --hash=sha256:a6df30f272c08bf8be66e0775fad862005d950a6b8449b94f7c788731d70ecd7
    # via -r requirements.in
tensorboard-data-server==0.7.2 \
    --hash=sha256:7e0610d205889588983836ec05dc098e80f97b7e7bbff7e994ebb78f578d0ddb \
    --hash=sha256:9fe5d24221b29625dbc7328b0436ca7fc1c23de4acf4d272f1180856e32f9f60 \

  // shape is explicitly requested where possible.
  int (*num_dims)(void* data, TF_Status* status);

  // Computes the axis length at `dim_index`.
  int64_t (*dim)(void* data, int dim_index, TF_Status* status);

  void (*deallocator)(void* data);

  // Summarizes the value of this tensor. The caller takes ownership of the

    ] + if_not_mobile([
        "//tensorflow/core/kernels:tensor_list",
        "//tensorflow/core/kernels:tensor_list_util",
        "//tensorflow/core/kernels:variant_ops_util",
        "//tensorflow/core/kernels/data:optional_ops_util",
        "//tensorflow/core/platform:abi",
    ]),
    alwayslink = 1,
)

tf_cuda_library(
    name = "ops",
    srcs = [
        "ops.cc",
    ],
    hdrs = [

*   `tf.data`:
    *   Graduated experimental APIs:
        * [`tf.data.Dataset.ragged_batch`](https://www.tensorflow.org/api_docs/python/tf/data/Dataset/#ragged_batch), which batches elements of `tf.data.Dataset`s into `tf.RaggedTensor`s.
        * [`tf.data.Dataset.sparse_batch`](https://www.tensorflow.org/api_docs/python/tf/data/Dataset/#sparse_batch), which batches elements of `tf.data.Dataset`s into `tf.sparse.SparseTensor`s.

  tensorflow::string value = cell->cell.value();
  void* data = tensorflow::port::Malloc(value.length());
  value.copy(static_cast<char*>(data), value.length(), 0);
  buf->data = data;
  buf->length = value.length();
  buf->data_deallocator = [](void* data, size_t length) {
    tensorflow::port::Free(data);
  };
}

TFE_MonitoringStringGauge0* TFE_MonitoringNewStringGauge0(