// CHECK-SAME: %arg21: tensor<!tf_type.resource<tensor<640xf32>>>
  // CHECK-SAME: %arg22: tensor<!tf_type.resource<tensor<640xf32>>>
  // CHECK-SAME: %arg23: tensor<!tf_type.resource<tensor<512x640xf32>>>
  // CHECK-SAME: %arg24: tensor<!tf_type.resource<tensor<640x640xf32>>>
  // CHECK-SAME: %arg25: tensor<!tf_type.resource<tensor<4096xf32>>>
  // CHECK-SAME: %arg26: tensor<!tf_type.resource<tensor<640x4096xf32>>>

    // CHECK:      metadata
    // CHECK-SAME: args
    // CHECK-SAME: dtype: DT_INT32
    // CHECK-SAME: shape
    // CHECK-SAME: dim
    // CHECK-SAME: size: 8
    // CHECK-SAME: kind: PARAMETER
    // CHECK-SAME: sharding
    // CHECK-SAME: type: MAXIMAL
    // CHECK-SAME: tile_assignment_dimensions: 1
    // CHECK-SAME: tile_assignment_devices: 0
    // CHECK-SAME: retvals
    // CHECK-SAME: sharding

class OperandsSameElementTypeConstraintBase<string op> :
  PredOpTrait<op # " operands have same element type",
    OperandsSameElementTypeConstraintBasePred>;

// This is a constraint for most of the binary ops, e.g., add, mul, div, etc.
// Binary ops lhs & rhs should have the same value type, and is capable to
// compare quantization types as well.
def BinaryOpSameElementTypeConstraint :

func.func @testHashTableAndInitializeTableToV2(%arg0: tensor<!tf_type.string>) {
  // CHECK: [[handle:%.*]] = "tf.HashTableV2"()
  // CHECK-SAME: container = ""
  // CHECK-SAME: key_dtype = !tf_type.string
  // CHECK-SAME: shared_name = "table"
  // CHECK-SAME: value_dtype = i32
  // CHECK-SAME: device = ""
  // CHECK-SAME: () -> tensor<!tf_type.resource>

}

// CHECK-LABEL: reduce_window_with_max
// CHECK-SAME: %[[ARG0:.*]]: tensor<2x9x10x3x!quant.uniform<i8:f32, 3.000000e-01:-5>>
// CHECK-SAME: %[[ARG1:.*]]: tensor<!quant.uniform<i8:f32, 3.000000e-01:-5>>
// CHECK: %[[MAX_POOL:.*]] = "tfl.max_pool_2d"(%[[ARG0]])

  // CHECK: "tfl.unidirectional_sequence_lstm"(%arg0,
  // CHECK-SAME: %arg1, %arg2, %arg3, %arg4, %arg5, %arg6, %arg7, %arg8, %arg9, %arg10, %arg11, %arg12, %arg13, %arg14, %arg15,
  // CHECK-SAME: %arg16, %arg17, %arg18, %arg19, %arg20, %arg21, %arg22, %arg23) <{cell_clip = 1.000000e+00 : f32, fused_activation_function = "NONE", time_major = false}> :
  // CHECK-SAME: (tensor<?x?x?xf32>,

<pre>
[...]Point{{1.5, -3.5}, {0, 0}}     // same as [...]Point{Point{1.5, -3.5}, Point{0, 0}}
[][]int{{1, 2, 3}, {4, 5}}          // same as [][]int{[]int{1, 2, 3}, []int{4, 5}}
[][]Point{{{0, 1}, {1, 2}}}         // same as [][]Point{[]Point{Point{0, 1}, Point{1, 2}}}
map[string]Point{"orig": {0, 0}}    // same as map[string]Point{"orig": Point{0, 0}}
map[Point]string{{0, 0}: "orig"}    // same as map[Point]string{Point{0, 0}: "orig"}

  auto device_attr = (*this)->getAttrOfType<StringAttr>("device");
  // Treat missing device attribute like unspecified (= empty string) attribute.
  // Note that different op instances with the same string (including empty
  // string) are seen as dependent (same resource instance).
  if (!device_attr) return "";
  return device_attr.str();
}

//===----------------------------------------------------------------------===//
// NotEqualOp

      return op.emitOpError("'input' and 'output' should have the same rank.");
    }

    // Check if input and output shapes are same
    for (int i = 0; i < input_type.getRank(); i++) {
      if (input_type.getDimSize(i) != output_type.getDimSize(i)) {
        return op.emitOpError(
            "'input' and 'output' should have the same shape.");
      }
    }
  }
  return success();
}

                should have all the fields, including the ones that have the same value
                as the default. This is different from `response_model_exclude_unset`
                in that if the fields are set but contain the same default values,
                they will be excluded from the response.