recurrent activation function for GRU from `hard_sigmoid` to `sigmoid`, and
    `reset_after` to True in 2.0. Historically recurrent activation is
    `hard_sigmoid` since it is fast than 'sigmoid'. With new unified backend
    between CPU and GPU mode, since the CuDNN kernel is using sigmoid, we change
    the default for CPU mode to sigmoid as well. With that, the default GRU will

The op returns an error if no system is running.
  }];

  let arguments = (ins);

  let results = (outs);
}

def TF_SigmoidOp : TF_Op<"Sigmoid", [Pure, TF_SameOperandsAndResultTypeResolveRef]> {
  let summary = "Computes sigmoid of `x` element-wise.";

  let description = [{
Specifically, `y = 1 / (1 + exp(-x))`.
  }];

  let arguments = (ins
    TF_FpOrComplexTensor:$x

A7D7          ; valid                                  # 14.0 LATIN SMALL LETTER MIDDLE SCOTS S
A7D8          ; mapped                 ; A7D9          # 14.0 LATIN CAPITAL LETTER SIGMOID S
A7D9          ; valid                                  # 14.0 LATIN SMALL LETTER SIGMOID S
A7DA..A7F1    ; disallowed                             # NA   <reserved-A7DA>..<reserved-A7F1>
A7F2          ; mapped                 ; 0063          # 14.0 MODIFIER LETTER CAPITAL C