// CHECK-NEXT:  [[HASH_TABLE:%.*]] = "tfl.hashtable"() <{key_dtype = !tf_type.string, table_id = 1530976467 : i32, value_dtype = i64}> : () -> tensor<1x!tf_type.resource>
  // CHECK-NEXT:  [[FIND:%.*]] = "tfl.hashtable_find"([[HASH_TABLE]], [[CST]], %arg0) : (tensor<1x!tf_type.resource>, tensor<!tf_type.string>, tensor<i64>) -> tensor<*xi64>
  // CHECK-NEXT:  return [[FIND]] : tensor<*xi64>
  func.return %1 : tensor<*xi64>
}

// -----

  CUMSUM = 128,
  CALL_ONCE = 129,
  BROADCAST_TO = 130,
  RFFT2D = 131,
  CONV_3D = 132,
  IMAG=133,
  REAL=134,
  COMPLEX_ABS=135,
  HASHTABLE = 136,
  HASHTABLE_FIND = 137,
  HASHTABLE_IMPORT = 138,
  HASHTABLE_SIZE = 139,
  REDUCE_ALL = 140,
  CONV_3D_TRANSPOSE = 141,
  VAR_HANDLE = 142,
  READ_VARIABLE = 143,
  ASSIGN_VARIABLE = 144,
}

  CUMSUM = 128,
  CALL_ONCE = 129,
  BROADCAST_TO = 130,
  RFFT2D = 131,
  CONV_3D = 132,
  IMAG=133,
  REAL=134,
  COMPLEX_ABS=135,
  HASHTABLE = 136,
  HASHTABLE_FIND = 137,
  HASHTABLE_IMPORT = 138,
  HASHTABLE_SIZE = 139,
  REDUCE_ALL = 140,
  CONV_3D_TRANSPOSE = 141,
  VAR_HANDLE = 142,
  READ_VARIABLE = 143,
  ASSIGN_VARIABLE = 144,
  BROADCAST_ARGS = 145,

    TypeAttr:$key_dtype,
    TypeAttr:$value_dtype
  );

  let results = (outs TFL_ResourceTensor:$out);

  let hasOptions = 1;
}

def TFL_HashtableFindOp: TFL_Op<"hashtable_find", []> {
  let summary = "Looks up keys in a table, outputs the corresponding values.";

  let description = [{
The tensor `keys` must of the same type as the keys of the table.

    "CUMSUM",
    "CALL_ONCE",
    "BROADCAST_TO",
    "RFFT2D",
    "CONV_3D",
    "IMAG",
    "REAL",
    "COMPLEX_ABS",
    "HASHTABLE",
    "HASHTABLE_FIND",
    "HASHTABLE_IMPORT",
    "HASHTABLE_SIZE",
    "REDUCE_ALL",
    "CONV_3D_TRANSPOSE",
    "VAR_HANDLE",
    "READ_VARIABLE",
    "ASSIGN_VARIABLE",
    "BROADCAST_ARGS",