//
  // For example some tensor handles may represent distributed values, in which
  // case placement information is lost when resolving the handle.
  //
  // If false, a caller might implement pretty-printing by resolving and
  // iterating over the resulting tensor. This may still be viable if resolving
  // the handle loses information, but `SummarizeValue` would be more precise.

// all be equal to `num_shapes_and_types`. If `ranks[i] != -1`, (i.e., if the
// rank is known), then it must be equal to the length of `shapes[i]`; if
// `ranks[i] == 1`, then `shapes[i]` may be nullptr.
//
// TODO(akshayka): Implement a corresponding getter method.
void TF_GraphSetOutputHandleShapesAndTypes(TF_Graph* graph, TF_Output output,
                                           int num_shapes_and_types,

// scope for now. I'm just mentioning this here as that transition will mean
// removal of the registration part from `Env` and adding it here instead: we
// will need tables to hold for each scheme the function tables that implement
// the needed functionality instead of the current `FileSystemRegistry` code in
// `core/platform/env.cc`.
class ModularFileSystem final : public FileSystem {
 public:
  ModularFileSystem(

unauthorized access to data or models, but also to prevent malicious users or
tenants sending graphs to execute under another tenant’s identity.

The isolation mechanisms are the responsibility of the users to design and
implement, and therefore security issues deriving from their absence are not
considered a vulnerability in TensorFlow.

### Resource allocation

A denial of service caused by one model could bring down the entire server, but

    to learn more.
*   Adds optional `exclude_cols` parameter to CsvDataset. This parameter is the
    complement of `select_cols`; at most one of these should be specified.
*   We have implemented an optimization which reorders data-discarding
    transformations such as `take` and `shard` to happen earlier in the dataset

  }
  const string& Name() const override { return op_type_; }
  const string& DeviceName() const override { return device_name_; }

  Status SetDeviceName(const char* name) override {
    // TODO(srbs): Implement this.
    device_name_ = name;
    return absl::OkStatus();
  }

  Status AddInput(AbstractTensorHandle* input) override {
    GraphTensor* t = dyn_cast<GraphTensor>(input);
    if (!t) {

  }

 private:
  TFE_Context* context_;
  TFE_CustomDevice device_;
  void* info_;
  string name_;
};

// An adapter which wraps the shape/data produced by C custom devices and uses
// it to implement custom device methods.
class CAPICustomDeviceTensorHandle
    : public tensorflow::CustomDeviceTensorHandle {
 public:
  CAPICustomDeviceTensorHandle(tensorflow::ImmediateExecutionContext* context,

          bucket.c_str(), object.c_str(), stat->base.length,
          stat->generation_number, stat->base.mtime_nsec);
  return TF_SetStatus(status, TF_OK, "");
}

// TODO(vnvo2409): Implement later
void NewRandomAccessFile(const TF_Filesystem* filesystem, const char* path,
                         TF_RandomAccessFile* file, TF_Status* status) {
  std::string bucket, object;

       ++component_index) {
    // TODO(allenl): Add a C API for summarizing tensors. Currently custom
    // devices limiting themselves to a C API (for ABI compatibility) would need
    // to implement summarization for component tensors themselves.
    ImmediateExecutionTensorHandle* component =
        tensorflow::unwrap(tensors_[component_index].get());
    std::string component_summary;

// operation whose operation type matches the function name.
typedef struct TF_Function TF_Function;

// Function definition options. TODO(iga): Define and implement
typedef struct TF_FunctionOptions TF_FunctionOptions;

// Sets the shape of the Tensor referenced by `output` in `graph` to
// the shape described by `dims` and `num_dims`.
//