Parallel training of recurrent networks is not supported. ''ExecutionEnvironment'' value in trainingOptions function must be ''auto'', ''gpu'', or ''cpu''


trainNetwork