learning and neural networks. However, the framework is versatile enough to be
used in other areas as well.

TensorFlow provides stable [Python](https://www.tensorflow.org/api_docs/python)
and [C++](https://www.tensorflow.org/api_docs/cc) APIs, as well as a
non-guaranteed backward compatible API for
[other languages](https://www.tensorflow.org/api_docs).

# Therefore, "tfrun" commands cannot include pipes -- which is
# probably for the better. If a pipe is necessary for something, it is probably
# complex. Write a well-documented script under utilities/ to encapsulate the
# functionality instead.
tfrun() { "$@"; }

# Run all "tfrun" commands under Docker. See setup_docker.sh for details
if [[ "$TFCI_DOCKER_ENABLE" == 1 ]]; then

#include "tensorflow/core/platform/test.h"

// NOTE(allenl): These tests currently go through TFE_Execute and so are
// integration testing rather than purely testing the parallel device. They
// correspond fairly well to the implementation, but testing the C++ directly is
// another option.

namespace tensorflow {
namespace parallel_device {

Variable* Variable::Create(TFE_Context* context, TF_DataType type,

#include "tensorflow/core/platform/test.h"

// NOTE(allenl): These tests currently go through TFE_Execute and so are
// integration testing rather than purely testing the parallel device. They
// correspond fairly well to the implementation, but testing the C++ directly is
// another option.

namespace tensorflow {
namespace parallel_device {

using ::testing::HasSubstr;

TEST(PARALLEL_DEVICE, TestBasicCPU) {