DataType dtype;
  TF_RETURN_IF_ERROR(ConvertToDataType(arg_type.getElementType(), &dtype));
  AttrValue type_attr;
  type_attr.set_type(dtype);
  (*node_def->mutable_attr())["T"] = type_attr;

  AttrValue index_attr;
  index_attr.set_i(index);
  (*node_def->mutable_attr())["index"] = index_attr;

  if (auto device_attr =

  DataType dtype;
  TF_RETURN_IF_ERROR(ConvertToDataType(arg_type.getElementType(), &dtype));
  AttrValue type_attr;
  type_attr.set_type(dtype);
  (*node_def->mutable_attr())["T"] = type_attr;

  AttrValue index_attr;
  index_attr.set_i(index);
  (*node_def->mutable_attr())["index"] = index_attr;

  if (auto device_attr =

            std::string type = "DT_UINT8";

            if (arg.type() != DT_INVALID) {
              type = DataType_Name(arg.type());
            } else if (!arg.type_attr().empty()) {
              OpDef_AttrDef attr =
                  *FindAttr(arg.type_attr(), op_info.graph_op_def);
              if (attr.has_default_value() &&
                  attr.default_value().value_case() == AttrValue::kType) {

  }
  return string(name);
}

void InferArgAttributes(const OpDef::ArgDef& arg,
                        std::unordered_map<string, string>* inferred_attrs) {
  if (!arg.type_attr().empty()) {
    gtl::InsertIfNotPresent(inferred_attrs, arg.type_attr(), arg.name());
  } else if (!arg.type_list_attr().empty()) {
    gtl::InsertIfNotPresent(inferred_attrs, arg.type_list_attr(), arg.name());
  }
  if (!arg.number_attr().empty()) {

        attr_type, attr_def.name, attr_def.type)
  else:
    attr_names = []
    if arg_def.number_attr:
      attr_names.append(arg_def.number_attr)
    if arg_def.type_attr:
      attr_names.append(arg_def.type_attr)
    if arg_def.type_list_attr:
      attr_names.append(arg_def.type_list_attr)

    # TODO(fengliuai): currently we don't support backward type inference, so we

    for (auto type_and_name :
         llvm::zip(intermediate_types, kIntermediateNames)) {
      mlir::TypeAttr type_attr =
          mlir::TypeAttr::get(std::get<0>(type_and_name));
      auto named_attr =
          builder.getNamedAttr(std::get<1>(type_and_name), type_attr);
      op_state.addAttribute(named_attr.getName(), named_attr.getValue());
    }
  }
  return absl::OkStatus();
}

      return false;
    }
  }
  return true;
}

// Returns the tensor type created from the `shape_attr` and `type_attr`
// attributes.
Type GetType(Attribute shape_attr, Attribute type_attr) {
  auto shape = mlir::cast<tf_type::ShapeAttr>(shape_attr);
  auto type = mlir::cast<TypeAttr>(type_attr);
  if (shape.hasRank())
    return tensorflow::GetTypeFromTFTensorShape(shape.getShape(),

    type_attr = attr_value_pb2.AttrValue(type=types_pb2.DT_QINT8)
    if quantize:
      self.assertTrue(
          self._contains_op(output_graphdef, 'Const', 'dtype', type_attr)
      )
    else:
      self.assertFalse(
          self._contains_op(output_graphdef, 'Const', 'dtype', type_attr)
      )

  @parameterized.named_parameters(

          name: "condition"
          type: DT_BOOL
        }
        input_arg {
          name: "t"
          type_attr: "T"
        }
        input_arg {
          name: "e"
          type_attr: "T"
        }
        output_arg {
          name: "output"
          type_attr: "T"
        }
        attr {
          name: "T"
          type: "type"
        }
      }
      op {

				if(!obj || obj === -1) { obj = this.get_container().find("> ul > li"); }
				li_attr = $.isArray(li_attr) ? li_attr : [ "id", "class" ];
				if(!is_callback && this.data.types) { li_attr.push(s.types.type_attr); }
				a_attr = $.isArray(a_attr) ? a_attr : [ ];

				obj.each(function () {
					li = $(this);
					tmp1 = { data : [] };
					if(li_attr.length) { tmp1.attr = { }; }