### Typical Pull Request Workflow -

**1. New PR**

- As a contributor, you submit a New PR on GitHub.
- We inspect every incoming PR and add certain labels to the PR such as `size:`,
  `comp:` etc.  At this stage we check if the PR is valid and meets certain
  quality requirements. For example, we check if the CLA is signed, PR has

// Fetches the current number of inputs attached to `op`.
//
// Does not use the operation's definition to determine how many inputs should
// be attached. It is intended for use with TFE_OpGetFlatInput to inspect an
// already-finalized operation.
//
// Note that TFE_OpGetFlatInputCount and TFE_OpGetFlatInput operate on a flat
// sequence of inputs, unlike TFE_OpGetInputLength (for getting the length of a

// APIs for generically dealing with op attributes (e.g. when forwarding them
// through custom device implementations).
//
// TODO(allenl): Currently these are black boxes, but we should have some way to
// inspect values. This would let people e.g. copy over most attributes and then
// modify some based on their values.

// A reference to an op's name -> attribute mapping
typedef struct TFE_OpAttrs TFE_OpAttrs;

  void DeleteTrace(int64_t tensor_id);

  // Consumes the internal state of the tape (so cannot be called more than
  // once) and produces the gradient of the target tensors with respect to the
  // source tensors. The output gradients are used if not empty and not
  // null. The result is populated with one tensor per target element.
  // When running backward functions, builds zeros-like tensors for

// ParallelDevice.
class ParallelTensor {
 public:
  // Construct a ParallelTensor from TensorHandles placed on the component
  // devices of a ParallelDevice. If called, ParallelTensor::Shape inspects
  // `components` to determine a shape.
  static std::unique_ptr<ParallelTensor> FromTensorHandles(
      const ParallelDevice& parallel_device,
      std::vector<TensorHandlePtr> components, TF_Status* status);