Results with Low Metrics

Hello,

I have deployed the CIFAR100 scenario (/OSR/OpenMax/cifar100.py) and I get about 62% of the F1 Micro metric. Isn't that value a bit low? Is there any scenario in this repository that gives better results?

Best regards,
Ander