node [fontname = "Sans"];
   edge [fontname = "Sans"];

   "apiElements" [shape="box", label=<<B>apiElements</B>(C)>, color="pink", bgcolor="pink", style="filled"]
   "apiElements" -> "api"
   "apiElements" -> "compileOnlyApi"

   "api" [shape="box", label=<<B>api</B>>, color="chartreuse3", bgcolor="chartreuse3", style="filled"]
   "compileOnlyApi" [shape="box", label=<<B>compileOnlyApi</B>>, color="chartreuse3", bgcolor="chartreuse3", style="filled"]

// CHECK-NEXT:     }, {
// CHECK-NEXT:       shape: [ 0 ],
// CHECK-NEXT:       name: "input_to_input_intermediate",
// CHECK-NEXT:       has_rank: true
// CHECK-NEXT:     }, {
// CHECK-NEXT:       shape: [ 0 ],
// CHECK-NEXT:       name: "input_to_forget_intermediate",
// CHECK-NEXT:       has_rank: true
// CHECK-NEXT:     }, {
// CHECK-NEXT:       shape: [ 0 ],
// CHECK-NEXT:       name: "input_to_cell_intermediate",

  auto c_api_client = down_cast<xla::PjRtCApiClient*>(pjrt_client->get());
  std::vector<int32_t> data(1, 0);
  xla::Shape shape = xla::ShapeUtil::MakeShape(xla::S32, {1});

  auto buffer = c_api_client->pjrt_c_client()->client->BufferFromHostBuffer(
      data.data(), shape.element_type(), shape.dimensions(),
      /*byte_strides=*/std::nullopt,
      xla::PjRtClient::HostBufferSemantics::kImmutableOnlyDuringCall, nullptr,

  %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