Christian S. Perone

@ranisalt if you think that a pull-request with 51 files changed, without testing implemented, will be merged fast, then you'll need to review what you think that an open-source contribution...

Done into the f84c12a394dce71b1449, still need testing, can @lorenzoriano test the MPIMigration again ?

There are a lot of tradeoffs in this, for instance, if you increase the number of individuals that are going to go to the next population you may also reduce...

There are a lot of tradeoffs in this, for instance, if you increase the number of individuals that are going to go to the next population you may also reduce...

Sure @mycaule ! This contribution is very welcome !

You're supposed to see always the same or better minimum raw score, not the fitness, note the difference.

Cool, I think that @alzaia also tried watershed. I personally think that adding separating weights in the loss during training (like they did on the original U-Net paper) would make...

What I mean is to use larger weights on the border of the myelin, like in these weights in the original u-net paper (they used morphology to compute it as...

Hi @patk-motional, can we reopen this issue ? It has been **more than 1 year** that the paper doesn't reflect the correct information about the framework and other datasets, that...

Thanks for the quick reply @patk-motional, this sounds fine. I think the type can also be Pred+Plan (also for nuPlan no ?) because it can be used for agent prediction...