%0 = "tf.Squeeze"(%arg0) : (tensor<1x2x2xf32>) -> tensor<2x2xf32>
  func.return %0 : tensor<2x2xf32>

// CHECK-LABEL:squeezeDefault
// CHECK:  "tfl.squeeze"(%arg0) <{squeeze_dims = []}> : (tensor<1x2x2xf32>) -> tensor<2x2xf32>
}

func.func @squeezeSingleAxis(%arg0: tensor<2x1x2xf32>) -> tensor<2x2xf32> {
  %0 = "tf.Squeeze"(%arg0) {squeeze_dims = [1]} : (tensor<2x1x2xf32>) -> tensor<2x2xf32>

// CHECK:           [[VAL_12:%.*]]:4 = "tf.SplitV"([[VAL_7]], [[VAL_10]], [[VAL_11]]) : (tensor<40x8xf32>, tensor<4xi32>, tensor<i32>) -> (tensor<10x8xf32>, tensor<10x8xf32>, tensor<10x8xf32>, tensor<10x8xf32>)
// CHECK-DAG:       [[VAL_13:%.*]] = "tf.Const"() <{value = dense<10> : tensor<4xi32>}> : () -> tensor<4xi32>

// -----

func.func @testDynamicStitch(%arg0: tensor<2x2xf32>) -> tensor<2x2xf32> {
  %indices = "tf.Const"() {value = dense<[1, 0]> : tensor<2xi32>} : () -> tensor<2xi32>
  %0 = "tf.DynamicStitch"(%indices, %arg0) : (tensor<2xi32>, tensor<2x2xf32>) -> tensor<2x2xf32>
  func.return %0 : tensor<2x2xf32>
}

// -----

func.func @testDynamicStitch() -> tensor<2x2xf32> {

// CHECK-LABEL: testFoldEnsureShapeOp
func.func @testFoldEnsureShapeOp(%arg0: tensor<10x20xf32>) -> (tensor<10x20xf32>, tensor<10x20xf32>, tensor<20x10xf32>) {
  %0 = "tf.EnsureShape"(%arg0) {shape = #tf_type.shape<10x20>} : (tensor<10x20xf32>) -> tensor<10x20xf32>
  %1 = "tf.EnsureShape"(%arg0) {shape = #tf_type.shape<?x20>} : (tensor<10x20xf32>) -> tensor<10x20xf32>
  // Failing case which should not be folded.

  %0 = stablehlo.uniform_dequantize %arg : (tensor<2x2x!quant.uniform<i8:f32, 1.000000e+0:8>>) -> tensor<2x2xf32>
  return %0 : tensor<2x2xf32>
}
// CHECK-LABEL: uniform_dequantize_op
// CHECK: %[[DEQUANT:.+]] = "tfl.dequantize"({{.*}}) : (tensor<2x2x!quant.uniform<i8:f32, 1.000000e+00:8>>) -> tensor<2x2xf32>
// CHECK: return %[[DEQUANT]]

// -----

        %6 = "tf.TensorListSetItem"(%3, %4, %5) {device = ""} : (tensor<!tf_type.variant<tensor<*xf32>>>, tensor<i32>, tensor<2x2xf32>)-> tensor<*x!tf_type.variant>
        %7 = "tf.Const"() {device = "", value = dense<-1> : tensor<i32>} : () -> tensor<i32>

}

// -----

func.func @add_with_quantized_i16_broadcasting(tensor<2x2xf32>, tensor<1xf32>) -> tensor<2x2x!quant.any<i16:f32>> {
^bb0(%arg0: tensor<2x2xf32>, %arg1: tensor<1xf32>):
  // expected-error @+1 {{Operands should have valid shapes and element type needs to match}}
  %0 = "tfl.add"(%arg0, %arg1) {fused_activation_function = "RELU6"} : (tensor<2x2xf32>, tensor<1xf32>) -> tensor<2x2x!quant.any<i16:f32>>

      // were:
      //   tensor<!tf_type.variant<tensor<?x8xf32>>>
      // and:
      //   tensor<!tf_type.variant<tensor<10x8xf32>>>
      // we'll try here to refine tensor<?x8xf32> with tensor<10x8xf32>.
      auto refined_subtype = mlir::cast<TensorType>(
          TypeMeet(lhs_element_type_with_subtype.GetSubtypes().front(),