Evaluating the Quality of HLS4ML’s Basic Neural Network Implementations on FPGAs