"""A model with 2 signatures.

  Used to test where the quantizer has to handle multiple signatures.
  """

  def __init__(self):
    self.matmul_filters = random_ops.random_uniform(
        shape=(4, 3), minval=-1.0, maxval=1.0
    )
    self.conv_filters = np.random.uniform(
        low=-10, high=10, size=(2, 3, 3, 2)
    ).astype('f4')

  @def_function.function(

}

TF_Operation* RandomUniform(TF_Operation* shape, TF_DataType dtype,
                            TF_Graph* graph, TF_Status* s) {
  TF_OperationDescription* desc =
      TF_NewOperation(graph, "RandomUniform", "random_uniform");
  TF_AddInput(desc, {shape, 0});
  TF_SetAttrType(desc, "dtype", dtype);
  return TF_FinishOperation(desc, s);
}

void Split3Helper(TF_Operation* input, TF_Graph* graph, TF_Status* s,

      output_tensor: The resulting tensor of the convolution operation.
    """
    in_placeholder = array_ops.placeholder(dtypes.float32, shape=input_shape)

    filters = random_ops.random_uniform(
        shape=filter_shape, minval=-1.0, maxval=1.0
    )
    if use_variable_for_filter:
      filters = variables.Variable(filters)

    output_tensor = nn_ops.conv2d(
        in_placeholder,

  REDUCE_ALL = 140,
  CONV_3D_TRANSPOSE = 141,
  VAR_HANDLE = 142,
  READ_VARIABLE = 143,
  ASSIGN_VARIABLE = 144,
  BROADCAST_ARGS = 145,
  RANDOM_STANDARD_NORMAL = 146,
  BUCKETIZE = 147,
  RANDOM_UNIFORM = 148,
  MULTINOMIAL = 149,
  GELU = 150,
  DYNAMIC_UPDATE_SLICE = 151,
  RELU_0_TO_1 = 152,
  UNSORTED_SEGMENT_PROD = 153,
  UNSORTED_SEGMENT_MAX = 154,
  UNSORTED_SEGMENT_SUM = 155,
  ATAN2 = 156,

}


// -----

// CHECK-LABEL: testRandomUniform
func.func @testRandomUniform(%arg0: tensor<3xi32>) -> tensor<?x?x?xf32> {
  // CHECK: "tfl.random_uniform"(%arg0)
  %0 = "tfl.random_uniform"(%arg0) {seed = 0 : i64, seed2 = 0 : i64} : (tensor<3xi32>) -> tensor<?x?x?xf32>
  func.return %0 : tensor<?x?x?xf32>
}

// -----

// CHECK-LABEL: testRandomStandardNormal

  %0 = "tf.RandomUniform"(%arg0) {seed = 0 : i64, seed2 = 0: i64} : (tensor<3xi32>) -> tensor<?x?x?xf32>
  func.return %0 : tensor<?x?x?xf32>

// CHECK-LABEL:random_uniform_f32
// CHECK:  "tfl.random_uniform"(%arg0) <{seed = 0 : i64, seed2 = 0 : i64}> : (tensor<3xi32>) -> tensor<?x?x?xf32>
}

func.func @random_standard_normal_f32(%arg0: tensor<3xi32>) -> tensor<?x?x?xf32> {

    F32ArrayAttr:$boundaries
  );

  let results = (outs
    TFL_TensorOf<[I32]>:$output
  );

  let hasOptions = 1;
}

def TFL_RandomUniformOp : TFL_Op<"random_uniform", []> {
  let summary = "Outputs random values from a uniform distribution.";

  let description = [{
The generated values follow a uniform distribution in the range `[0, 1)`. The

    "REDUCE_ALL",
    "CONV_3D_TRANSPOSE",
    "VAR_HANDLE",
    "READ_VARIABLE",
    "ASSIGN_VARIABLE",
    "BROADCAST_ARGS",
    "RANDOM_STANDARD_NORMAL",
    "BUCKETIZE",
    "RANDOM_UNIFORM",
    "MULTINOMIAL",
    "GELU",
    "DYNAMIC_UPDATE_SLICE",
    "RELU_0_TO_1",
    "UNSORTED_SEGMENT_PROD",
    "UNSORTED_SEGMENT_MAX",
    "UNSORTED_SEGMENT_SUM",
    "ATAN2",

        operations to delegate instead of SDK version.
*   `tf.random`:
    *   Various random number generation improvements:
    *   Add a fast path for default `random_uniform`
    *   `random_seed` documentation improvement.
    *   `RandomBinomial` broadcasts and appends the sample shape to the left
        rather than the right.