Args:
          use_variable: If True, creates a variable for weight.
        """
        super().__init__()
        w_val = np.random.randn(128, 32).astype('f4')
        if use_variable:
          self.w = variables.Variable(w_val)
        else:
          self.w = w_val

      @def_function.function(
          input_signature=[
              tensor_spec.TensorSpec(

// Replace
//   (float)OneHot(index, depth, on_val, off_val, axis)
// With
//   OneHot(index, depth, (float)on_val, (float)off_val, axis)
def FuseOneHotAndCastToFloat : Pat<
  (TFL_CastOp:$output (TFL_OneHotOp $indices,
                                    $depth,
                                    (Arith_ConstantOp $on_val),
                                    (Arith_ConstantOp $off_val),

          use_variable: If True, creates a variable for weight.
        """
        super(GatherModel, self).__init__()
        w_val = np.random.randn(128, 32).astype('f4')
        if use_variable:
          self.w = variables.Variable(w_val)
        else:
          self.w = w_val

      @def_function.function(
          input_signature=[
              tensor_spec.TensorSpec(

		{"FreeAddrInfoW", Func, 1},
		{"FreeEnvironmentStrings", Func, 0},
		{"FreeLibrary", Func, 0},
		{"Fsid", Type, 0},
		{"Fsid.Val", Field, 0},
		{"Fsid.X__fsid_val", Field, 2},
		{"Fsid.X__val", Field, 0},
		{"Fstat", Func, 0},
		{"Fstatat", Func, 12},
		{"Fstatfs", Func, 0},
		{"Fstore_t", Type, 0},
		{"Fstore_t.Bytesalloc", Field, 0},
		{"Fstore_t.Flags", Field, 0},

  auto _outputs = _o->outputs.size() ? _fbb.CreateVector<::flatbuffers::Offset<tflite::TensorMap>> (_o->outputs.size(), [](size_t i, _VectorArgs *__va) { return CreateTensorMap(*__va->__fbb, __va->__o->outputs[i].get(), __va->__rehasher); }, &_va ) : 0;

    // number without knowing this, so we need to manually dispatch to each
    // possible member of TensorProto, depening on its dtype.
#define MATCH(DTYPE, FIELD) \
  case DTYPE:               \
    return tensor.FIELD##_val().size()

  switch (tensor.dtype()) {
    MATCH(DT_FLOAT, float);
    MATCH(DT_DOUBLE, double);
    MATCH(DT_INT8, int);
    MATCH(DT_UINT8, int);
    MATCH(DT_INT16, int);
    MATCH(DT_UINT16, int);

  if (shape_x.size() == shape_y.size()) {
    llvm::SmallVector<int64_t, 4> out_shape(shape_x.size());
    for (int i = 0; i < shape_x.size(); i++) {
      auto x_val = shape_x[i];
      auto y_val = shape_y[i];
      out_shape[i] = std::max(x_val, y_val);
    }
    return tensorflow::GetTypeFromTFTensorShape(out_shape, element_type);
  }

  auto shape_large = shape_x.size() > shape_y.size() ? shape_x : shape_y;

	DT_ENCODING     DynTag = 32 /* Values greater than or equal to DT_ENCODING
	   and less than DT_LOOS follow the rules for
	   the interpretation of the d_un union
	   as follows: even == 'd_ptr', even == 'd_val'
	   or none */
	DT_PREINIT_ARRAY   DynTag = 32 /* Address of the array of pointers to pre-initialization functions. */
	DT_PREINIT_ARRAYSZ DynTag = 33 /* Size in bytes of the array of pre-initialization functions. */