)
    # Insert fake quant to simulate a QAT model.
    weight = array_ops.fake_quant_with_min_max_args(
        weight, min=-0.1, max=0.2, num_bits=8, narrow_range=False
    )

    # shape: (2, 2)
    output_tensor = math_ops.matmul(matmul_input, weight)
    # Insert fake quant to simulate a QAT model.
    output_tensor = array_ops.fake_quant_with_min_max_args(
        output_tensor, min=-0.2, max=0.2, num_bits=8, narrow_range=False

  }];

  let description = [{
    Given a const min, max, num_bits and narrow_range attribute, applies the
    same uniform quantization simulation as is done by the TensorFlow
    fake_quant_with_min_max_args op. See the fakeQuantAttrsToType() utility
    method and the quant-convert-simulated-quantization pass for further details.
  }];

  let arguments = (ins
    F32Tensor:$inputs,
    F32Attr:$min,

  }];

  let description = [{
    Given a const min, max, num_bits and narrow_range attribute, applies the
    same uniform quantization simulation as is done by the TensorFlow
    fake_quant_with_min_max_args op. See the fakeQuantAttrsToType() utility
    method and the quant-convert-simulated-quantization pass for further details.
  }];

  let arguments = (ins
    F32Tensor:$inputs,
    F32Attr:$min,

      def _matmul(self, x, y):
        x = array_ops.fake_quant_with_min_max_vars(
            x,
            min=ops.convert_to_tensor(self._min[0]),
            max=ops.convert_to_tensor(self._max[0]),
            num_bits=8,
            narrow_range=False,
        )
        y = array_ops.fake_quant_with_min_max_vars(
            y,
            min=ops.convert_to_tensor(self._min[1]),

  // CHECK: return [[VAL8]] : tensor<1x8x2xf32>
  func.return %2 : tensor<1x8x2xf32>
}

func.func @fake_quant_with_min_max_args(%arg0 : tensor<?x?xf32>) -> tensor<?x?xf32> {
  // CHECK-DAG: [[VAL0:%.+]] = "tf.Const"() <{value = dense<1.275000e+02> : tensor<f32>}>
  // CHECK-DAG: [[VAL1:%.+]] = "tf.Const"() <{value = dense<1.00392163> : tensor<f32>}>

// RUN: odml-to-stablehlo-opt %s --tf-stablehlo=skip-quantization-ops=false | FileCheck %s --check-prefix=CHECK-NOSKIP

func.func @fake_quant_with_min_max_vars(%arg0: tensor<1x1x28x48xf32>, %arg1: tensor<f32>, %arg2: tensor<f32>) -> tensor<1x1x28x48xf32> {

        `tf.math.zeta`.
    *   New endpoints in `tf.quantization` namespace:
        `tf.quantization.dequantize`,
        `tf.quantization.fake_quant_with_min_max_args`,
        `tf.quantization.fake_quant_with_min_max_args_gradient`,
        `tf.quantization.fake_quant_with_min_max_vars`,
        `tf.quantization.fake_quant_with_min_max_vars_gradient`,
        `tf.quantization.fake_quant_with_min_max_vars_per_channel`,

`min_adj = scale * round(min / scale)` and `max_adj = max + min_adj - min`.


Examples

```python

inp = tf.constant ([10.03, -10.23, 3])
out = tf.quantization.fake_quant_with_min_max_args(inp, min=-5, max=5,
                                                   num_bits=16)
print(out)

#  Output:
#  tf.Tensor([ 4.9999237 -5.0000763  3.0000763], shape=(3,), dtype=float32)
```