%0 = "tf.AnonymousIteratorV3"() {output_shapes = [
    #tf_type.shape<200x28x28x1>,
    #tf_type.shape<200x10>], output_types = [f32, f32]} : () -> tensor<!tf_type.resource>
  // CHECK: "tf.Iterator"
  // CHECK-SAME: shared_name = "_iterator2"
  %1 = "tf.AnonymousIteratorV3"() {output_shapes = [
    #tf_type.shape<200x10>], output_types = [f32]} : () -> tensor<!tf_type.resource>
  return
}

// -----

  return %arg4 : tensor<10xf32>
}

// -----

// CHECK-LABEL: strips_attributes_with_tf_values
// CHECK-NOT: tf
func.func @strips_attributes_with_tf_values()
          attributes {foo = #tf_type.shape<32x28x28x1>} {
  return
}

// -----

// CHECK-NEXT:   subgraphs: [ {
// CHECK-NEXT:     tensors: [ {
// CHECK-NEXT:       shape: [ 1, 4 ],
// CHECK-NEXT:       buffer: 1,
// CHECK-NEXT:       name: "arg0",
// CHECK-NEXT:       quantization: {
// CHECK-EMPTY:
// CHECK-NEXT:       },
// CHECK-NEXT:       has_rank: true
// CHECK-NEXT:     }, {
// CHECK-NEXT:       shape: [ 4, 4 ],
// CHECK-NEXT:       buffer: 2,
// CHECK-NEXT:       name: "arg1",

// CHECK-NEXT:   subgraphs: [ {
// CHECK-NEXT:     tensors: [ {
// CHECK-NEXT:       shape: [ 1, 4 ],
// CHECK-NEXT:       buffer: 1,
// CHECK-NEXT:       name: "arg0",
// CHECK-NEXT:       quantization: {
// CHECK-EMPTY:
// CHECK-NEXT:       },
// CHECK-NEXT:       has_rank: true
// CHECK-NEXT:     }, {
// CHECK-NEXT:       shape: [ 4, 4 ],
// CHECK-NEXT:       buffer: 2,
// CHECK-NEXT:       name: "arg1",

  return transposed_type;
}

// Returns shape of a ranked tensor.
// Precondition: output_val's is ranked tensor.
// Returns a truncated shape when `truncate` is set to true.
inline DenseElementsAttr GetShape(Value output_val, bool truncate = false) {
  auto output_shape = output_val.getType().dyn_cast<ShapedType>().getShape();

  SmallVector<int32_t> shape;
  shape.reserve(output_shape.size());

          "Shape must be fully-defined, got: shape[", dim, "] = ", shape[dim]));
    }
    num_elements *= shape[dim];
  }
  if (data.size() != num_elements) {
    return tensorflow::errors::InvalidArgument(absl::StrCat(
        "Mismatched shape and data size: \n", "Shape num_elements: ",
        num_elements, "\n", "Data size: ", data.size(), "\n"));
  }

import gradlebuild.basics.decapitalize
import gradlebuild.shade.ArtifactTypes.buildReceiptType
import gradlebuild.shade.ArtifactTypes.classTreesType
import gradlebuild.shade.ArtifactTypes.entryPointsType
import gradlebuild.shade.ArtifactTypes.manifestsType
import gradlebuild.shade.ArtifactTypes.relocatedClassesAndAnalysisType
import gradlebuild.shade.ArtifactTypes.relocatedClassesType
import gradlebuild.shade.extension.ShadedJarExtension

// Extracts shape from XlaArgument as TensorShape. If shape is a xla::Shape,
// that is converted to a TensorShape.
absl::StatusOr<TensorShape> GetTensorShapeFromXlaArgument(
    const XlaArgument& arg) {
  if (absl::holds_alternative<xla::Shape>(arg.shape)) {
    TensorShape arg_shape;
    TF_RETURN_IF_ERROR(
        XLAShapeToTensorShape(std::get<xla::Shape>(arg.shape), &arg_shape));
    return arg_shape;

  if (input_type_size < output_type_size) {
    TF_ShapeInferenceContextWithRankAtLeast(ctx, shape, 1, shape, status);

    if (TF_GetCode(status) == TF_OK) {
      TF_DimensionHandle* last_dim = TF_NewDimensionHandle();
      size_t divisor_val = output_type_size / input_type_size;
      TF_ShapeInferenceContextDim(ctx, shape, -1, last_dim);
      if (!TF_DimensionHandleValueKnown(last_dim) ||

  dltensor_in->device = {kDLCPU, 0};
  dltensor_in->ndim = static_cast<int32_t>(shape.size());
  dltensor_in->dtype = {kDLFloat, 32, 1};
  dltensor_in->shape = shape.data();
  dltensor_in->strides = strides.data();
  TFE_TensorHandle* handle = TFE_HandleFromDLPack(&dlm_in, status, ctx);
  ASSERT_NE(handle, nullptr)
      << TF_Message(status) << " (shape=[" << absl::StrJoin(shape, ",")
      << "], strides=[" << absl::StrJoin(strides, ",") << "])";