The path to the dataset in the file _code/config/tt100kclassif.py is wrong and raises an error when trying to train with this dataset. We have noticed a couple of teams have changed it in their repositories, but maybe it should be fixed in the base repository as it will affect everyone? (Another option is to change the name of the folder in the cluster)
The path to the dataset in the file _code/config/tt100kclassif.py is wrong and raises an error when trying to train with this dataset. We have noticed a couple of teams have changed it in their repositories, but maybe it should be fixed in the base repository as it will affect everyone? (Another option is to change the name of the folder in the cluster)