Diviyan Kalainathan
Diviyan Kalainathan
Hello @NawTun, It depends on the algorithm, but currently `.predict_proba` outputs the raw output of the algorithm. Which one are you using ? Best, Diviyan
Yes that is correct. The algorithm is more confident in a -> b than c ->d. Did you check with the original (matlab) implementation of ANM ? Those values does...
Hello ! Whoops ! Yes we should do this step before, thanks for the feedback !
Hi, The code is actually inspired from the author's code actually ; the convolutions are actually the embedding layers (they are not Linear layers), check on the end of p.4...
Hi again, I do understand your issue; I actually had the same issue, before convincing myself that it was the correct thing. I'll get into it next week, and I'll...
I've asked the author about it, I hope he gets back to me
Hi, I haven't gotten any response from the original author. I will dig into the algorithm next month to try close this issue. Feel free to make a pull request...
Hello, does this refer in some way to #14 ? I don't want to change twice the sign of the score, unless I was really wrong Best, Diviyan
This is strange ; It might directly come from the CAM package. Can you try running the same algorithm on a sample dataset (cdt.data.load_dataset('sachs') for example? ) ? Thanks a...
Hello, I think it comes from the CAM R package... This might take some time to fix ; as it will come with the recoding of all the R packages...