You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
由于一些跨域推荐模型采用的leave-one-out data splitting策略,但是在Recbole-CDR项目中具体执行时,评估设置如下:
eval_args:
split: { 'LS': 'valid_and_test' } #leave-one-out data splitting
group_by: user
order: RO
mode: uni999
报错: python3.7/site-packages/recbole/data/interaction.py",line367,in<dictcomp> new_inter = {col: torch.cat ([inter[col) for inter in interactions]) for col in columns_set} TypeError: expected Tensor as element 0 in argument 0, but got Interaction
The text was updated successfully, but these errors were encountered:
由于一些跨域推荐模型采用的leave-one-out data splitting策略,但是在Recbole-CDR项目中具体执行时,评估设置如下:
报错:
python3.7/site-packages/recbole/data/interaction.py",line367,in<dictcomp> new_inter = {col: torch.cat ([inter[col) for inter in interactions]) for col in columns_set} TypeError: expected Tensor as element 0 in argument 0, but got Interaction
The text was updated successfully, but these errors were encountered: