#include "tensorflow/c/conversion_macros.h"
#include "tensorflow/c/eager/abstract_op_attrs.h"
#include "tensorflow/c/tf_status.h"
#include "tensorflow/core/framework/attr_value.pb.h"

// An equivalent of a tensorflow::NameAttrList protocol buffer, but used in ways
// that sometimes do not require serialization.
typedef struct TFE_OpAttrs TFE_OpAttrs;

typedef struct TFE_Context TFE_Context;
typedef struct TFE_Op TFE_Op;

// Sets the attribute attr_name to be a function specified by 'function'.
//
// TODO(ashankar,iga): Add this functionality to the C API for graph
// construction. Perhaps we want an AttrValueMap equivalent in the C API?
TF_CAPI_EXPORT extern void TFE_OpSetAttrFunction(TFE_Op* op,
                                                 const char* attr_name,
                                                 const TFE_Op* value);

# variables in it, OR may be left empty if the user has already exported the
# relevant variables in their environment. Because of 'set -o allexport' above
# (which is equivalent to "set -a"), every variable in the file is exported
# for other files to use.
#
# Separately, if TFCI is set *and* there are also additional TFCI_ variables

  if (!status->status.ok()) return nullptr;

  // Create a TF_Graph from the MetaGraphDef. This is safe as long as Session
  // extends using GraphDefs. The Graph instance is different, but equivalent
  // to the one used to create the session.
  //
  // TODO(jhseu): When Session is modified to take Graphs instead of
  // GraphDefs, return the Graph generated in LoadSavedModel().

}  // namespace

// Allows a single op at a time to be launched without blocking.
//
// DeviceThread itself is thread-safe, in that StartExecute will block if there
// is a pending execution. Since StartExecute is equivalent to grabbing a lock,
// multiple DeviceThreads should always be accessed in the same order to avoid
// deadlocks.
class DeviceThread {
 public:
  // Starts a background thread waiting for `StartExecute`.

                                   absl::Span<AbstractTensorHandle*>)>;

// Runs `model` maybe wrapped in a function call op. This can be thought as
// being equivalent to the following python code.
//
// if use_function:
//   outputs = tf.function(model)(inputs)
// else:
//   outputs = model(inputs)
Status RunModel(Model model, AbstractContext* ctx,

computation
[**graphs**](https://developers.google.com/machine-learning/glossary/#graph).
Since models are practically programs that TensorFlow executes, using untrusted
models or graphs is equivalent to running untrusted code.

If you need to run untrusted models, execute them inside a
[**sandbox**](https://developers.google.com/code-sandboxing). Memory corruptions

    const std::function<void(BackwardFunction*)>& backward_function_deleter,
    const std::vector<Gradient*>& in_grads, absl::Span<Gradient*> out_grads) {
  /* This function is approximately equivalent to this Python code:

  forwardprop_aids = tf.ones_like(output_tensors)
  with tf.GradientTape() as g:
    g.watch(forwardprop_aids)
    grad = backward_function(forwardprop_aids)

    except NameError:
      answer = input(question)  # pylint: disable=bad-builtin
  except EOFError:
    answer = ''
  return answer


def symlink_force(target, link_name):
  """Force symlink, equivalent of 'ln -sf'.

  Args:
    target: items to link to.
    link_name: name of the link.
  """
  try:
    os.symlink(target, link_name)
  except OSError as e:
    if e.errno == errno.EEXIST:

  ///
  /// If `status` is not `TF_OK`, returns `false`, otherwise returns the same
  /// as the `is_directory` member of a `TF_FileStatistics` that would be used
  /// on the equivalent call of `stat`.
  ///
  /// Plugins:
  ///   * Must set `status` to `TF_OK` if `path` exists.
  ///   * Must set `status` to `TF_NOT_FOUND` if `path` doesn't point to a
  ///     filesystem entry.