// CHECK-NOT: func private @internal_quantize_i8
// CHECK-NOT: func private @internal_matmul_fn
// CHECK: func private @quantized_matmul_fn
// CHECK-SAME: tf_quant.quantized_ops = ["MatMul"]
// CHECK: func private @quantized_conv2d_fn
// CHECK-SAME: tf_quant.quantized_ops = ["Conv2D"]
// CHECK: func private @quantized_depthwise_conv2d_fn
// CHECK-SAME: tf_quant.quantized_ops = ["DepthwiseConv2D"]

 ArgMax 1
 AssignAddVariableOp 1
 BiasAdd 1
 BiasAddGrad 1
 Cast 115
 Const 407
 Conv2D 53
 Conv2DBackpropFilter 53
 Conv2DBackpropInput 52
 Equal 1
 FusedBatchNormGradV2 53
 FusedBatchNormV2 53
 MatMul 3
 MaxPool 1
 MaxPoolGrad 1
 Mean 1
 Mul 218
 Pad 2
 ReadVariableOp 538
 Relu 49
 ReluGrad 49
 Reshape 2
 ResourceApplyKerasMomentum 161
 Slice 1
 Softmax 1

  // CHECK: %[[TRANSPOSE:.*]] = "tf.Transpose"(%[[DEQUANT]], %[[CST]]) : (tensor<3x4xf32>, tensor<?xi32>) -> tensor<*xf32>
  // CHECK: %[[MATMUL:.*]] = "tf.MatMul"(%arg0, %[[TRANSPOSE]]) <{grad_a = false, grad_b = false, transpose_a = false, transpose_b = true}> : (tensor<2x3xf32>, tensor<*xf32>) -> tensor<2x4xf32>
  // CHECK: return %[[MATMUL]] : tensor<2x4xf32>
}

  %x = "tf.TensorArrayReadV3"(%handle_0, %index, %flow_0) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<2x!tf_type.resource<tensor<?x100xf32>>>, tensor<i32>, tensor<f32>) -> tensor<?x100xf32>
  %y = "tf.MatMul"(%x, %cst) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<?x100xf32>, tensor<100x512xf32>) -> (tensor<?x512xf32>)

// RUN: tf-quant-opt %s -split-input-file -quant-add-dump-tensor-op='debugger_type=float_per_layer' | FileCheck --check-prefix=FloatPerLayer %s

module {
  func.func @matmul2(%arg0: tensor<?x2xf32> {tf_saved_model.index_path = ["input_tensor"]}) -> (tensor<?x2xf32>) {
    %0 = stablehlo.constant dense<[-0.211145893, -0.708605706]> : tensor<2xf32>

  string output_dir = "tensorflow/c/experimental/ops/gen/cpp/golden";
  string source_dir = "tensorflow";
  string api_dirs = "";
  std::vector<string> ops = {
      "Neg",        // Simple unary Op
      "MatMul",     // 2 inputs & attrs with default values
      "IdentityN",  // Variadic input+output
      "SparseSoftmaxCrossEntropyWithLogits",  // 2 outputs
      "AccumulatorApplyGradient",             // 0 outputs

namespace tensorflow {
namespace ops {

//
Status Neg(AbstractContext* ctx, AbstractTensorHandle* const x, AbstractTensorHandle** y, const char* name = nullptr, const char* raw_device_name = nullptr);

//

def Test():

  x = tf.constant([[1.0], [1.0], [1.0]])
  y = tf.compat.v1.get_variable(
      name='y',
      shape=(1, 3),
      initializer=tf.random_normal_initializer(),
      trainable=True)
  r = tf.matmul(x, y)

  tensor_info_x = tf.compat.v1.saved_model.utils.build_tensor_info(x)
  tensor_info_r = tf.compat.v1.saved_model.utils.build_tensor_info(r)

  return {

  void TestMatMulGrad(const bool t_x, const bool t_y) {
    TestMatMulGradHelper<T>(
        /*is_x_batch=*/false, /*is_y_batch=*/false, t_x, t_y,
        [&](Output x, Output y) {
          return MatMul(root_, x, y, MatMul::TransposeA(t_x).TransposeB(t_y));
        });
  }

  template <typename T>
  void TestBatchMatMulGrad(const bool t_x, const bool t_y) {
    TestMatMulGradHelper<T>(

  auto matmul = rewriter.create<TFL::BatchMatMulOp>(
      loc, RankedTensorType::get(matmul_shape, result_type.getElementType()),
      lhs_flattend, rhs_flattend, /*adj_x*/ false_attr, /*adj_y*/ false_attr,
      /*asym_quant_input*/ false_attr);
  if (result_type.hasStaticShape()) {
    auto reshaped =
        rewriter.create<mhlo::ReshapeOp>(loc, result_type, matmul.getResult());
    return reshaped.getResult();