// model.
table SubGraph {
  // A list of all tensors used in this subgraph.
  tensors:[Tensor];

  // Indices of the tensors that are inputs into this subgraph. Note this is
  // the list of non-static tensors that feed into the subgraph for inference.
  inputs:[int];

  // Indices of the tensors that are outputs out of this subgraph. Note this is
  // the list of output tensors that are considered the product of the

#include "absl/container/flat_hash_map.h"
#include "tensorflow/core/graph/tensor_id.h"

namespace tensorflow {
namespace deadness_analysis_internal {

// Returns a map describing the predicate each Tensor was mapped to.  For
// testing purposes only.
using PredicateMapTy = absl::flat_hash_map<TensorId, string, TensorId::Hasher>;
Status ComputePredicates(const Graph& graph, PredicateMapTy* out_predicate_map,

    auto* ret = emptyTensor->tensor;
    delete emptyTensor;
    return ret;
  }

  Tensor tensor;
  if (!tensor.CopyFrom(src, src.shape())) {
    return nullptr;
  }
  return new tensorflow::TensorInterface(std::move(tensor));
}

// Non-static for testing.
TF_Tensor* TF_TensorFromTensor(const tensorflow::Tensor& src, Status* status) {

  }];

  let arguments = (ins
    Arg<TF_Int32Tensor, [{int32 `Tensor`.}]>:$arr,
    Arg<TF_Int32Tensor, [{non-negative int32 scalar `Tensor`.}]>:$size,
    Arg<TensorOf<[TF_Float32, TF_Float64, TF_Int32, TF_Int64]>, [{is an int32, int64, float32, or float64 `Tensor` with the same
shape as `arr`, or a length-0 `Tensor`, in which case it acts as all weights
equal to 1.}]>:$weights
  );

  TF_ASSERT_OK_AND_ASSIGN(Tensor tensor1,
                          runtime.CreateHostTensor<float>({}, TF_FLOAT, {2.0f}))
  TF_ASSERT_OK_AND_ASSIGN(Tensor tensor2,
                          runtime.CreateHostTensor<float>({}, TF_FLOAT, {3.0f}))

  TF_ASSERT_OK_AND_ASSIGN(Tensor result_tensor,
                          fn.Call<Tensor>(tensor1, tensor2));
  float out_val[1];

  for (int component_index = 0; component_index < tensors_.size();
       ++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());

  DataType data_type() const { return data_type_; }
  Status status() const { return status_; }
  const Tensor& tensor() const { return tensor_; }

 private:
  Status status_;
  Output output_ = Output(Operation(nullptr), 0);
  Tensor tensor_;
  const std::string node_name_ = "";
  int32 index_ = 0;
  DataType data_type_ = DT_INVALID;
};

// RUN: tf-opt "-tfl-lower-static-tensor-list=allow-tensorlist-pass-through default-to-single-batch=false" -split-input-file %s | FileCheck %s

// -----

func.func @tensorlistReserveConstantUnknownElementShapeDim(%arg0: tensor<i32>, %arg1: tensor<i32>) -> tensor<?x7xf32> {
  %cst = arith.constant dense<[-1, 7]> : tensor<2xi32>
  %0 = "tf.TensorListReserve"(%cst, %arg0) : (tensor<2xi32>, tensor<i32>) -> tensor<!tf_type.variant<tensor<?x7xf32>>>

//
// Conditions for `tfl.fully_connected` conversion:
//   * Input tensors are per-tensor uniform quantized (i8->f32)
//     tensors.
//   * The filter tensor is constant a per-tensor uniform quantized (i8->f32)
//     tensor. The quantization dimension should be 1 (the non-contracting
//     dimension).
//   * Output tensors are per-tensor uniform quantized (i8->f32) or
//     per-channel uniform quantized (i32->f32) tensors.

          (TF_SubOp $max_range, $min_range),
          (TF_ConstOp (GetF32Scalar<255>)))),
      $min_range),
    /*truncate=*/ConstBoolAttrFalse),
  [(TensorOf<[TF_Qint8, TF_Quint8]> $input),
   (TF_Float32Tensor $min_range),
   (TF_Float32Tensor $max_range),
   (TensorOf<[TF_Float32, TF_Bfloat16]> $result)]>;

//===----------------------------------------------------------------------===//
// Difference op patterns.