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
When I try to run the PD1 benchmark cifar100_wideresnet_2048, I get the error that sklearn is required for xgboost. After installing scikit-learn manually, everything works fine. It is simply missing in the dependencies or I messed something up during installation.
Steps to reproduce/How I installed MFPBench:
Setup new conda env: conda create --name mfp_test python=3.8
When I try to run the PD1 benchmark
cifar100_wideresnet_2048
, I get the error thatsklearn
is required forxgboost
. After installingscikit-learn
manually, everything works fine. It is simply missing in the dependencies or I messed something up during installation.Steps to reproduce/How I installed MFPBench:
conda create --name mfp_test python=3.8
conda activate mfp_test
mf-prior-bench
:pip install mf-prior-bench[pd1]
Now when querying the benchmark it returns the error
ImportError: sklearn needs to be installed in order to use this module
.After manually installing via
pip install scikit-learn
everything works fine.Please let me know if I messed something up when setting up mfpbench :)
The text was updated successfully, but these errors were encountered: