Diviyan Kalainathan
Diviyan Kalainathan
Which package version are you running ?
Could you update your package to the latest version (0.5.21) and try again ?
My guess is that 50 pairs for training NCC is nowhere near enough, i would suggest using a polynomial generator to generate ~2000 pairs. ``` from cdt.data import CausalPairGenerator c...
> When 50 testing vs 50 training, if epochs=500, average acc is ~50% as mentioned, while if epochs=200, average acc goes up to ~55%. There might be some overfitting hidden...
Hi, Right, I'll look into it.
Hello, I didn't get an answer from the author, I will get back to the implementation myself.
Hello ! This error stems from the docker image that does not possess your /etc/group and /etc/passwd (as it is personal and configuration-dependent). A way to solve your error message...
Ah yes, I forgot to recreate the link in the docker (we don't have a .bashrc, so the path links are not there) To have a python console, run `usr/bin/python3`,...
Ah right ! Sorry, I think we need to create a new Docker image ; / ## Dockerfile ```dockerfile FROM fentechai/cdt:latest RUN /usr/bin/python3 -m pip install ipython CMD /bin/bash ```...
Hello, You are right, I should be a bit more explicit on this one: for the pairwise models, we are using the Kaggle Cause-Effect Pair Challenge format [(link here)](https://www.kaggle.com/c/cause-effect-pairs/overview): [Here...