Open puhsu opened 1 year ago
I'm reading the dataset code and there is probably a bug here:
https://github.com/ml-jku/hopular/blob/3e0c39fdc59568349373af573ee52c03305ca105/hopular/auxiliary/data.py#L597-L602
After that, you are using the old indices to index into the shuffled/concatenated arrays. So the splits are different (not stratified, for example):
https://github.com/ml-jku/hopular/blob/3e0c39fdc59568349373af573ee52c03305ca105/hopular/auxiliary/data.py#L637-L638
I'm reading the dataset code and there is probably a bug here:
https://github.com/ml-jku/hopular/blob/3e0c39fdc59568349373af573ee52c03305ca105/hopular/auxiliary/data.py#L597-L602
After that, you are using the old indices to index into the shuffled/concatenated arrays. So the splits are different (not stratified, for example):
https://github.com/ml-jku/hopular/blob/3e0c39fdc59568349373af573ee52c03305ca105/hopular/auxiliary/data.py#L637-L638