I tried the API get_conditional_probs() that returns the list of CP for each LF. If my understanding is right, result returns the conditional probability when K =-1 (abstain) and Y=0 (in binary category in my case) . Could you explain how to use the weights to evaluate LF?…
How to determine the weights of LFs?
When I applied model's get_weights() API the results show some LFs are weighted as 0 and one LF is weighted as high as .93 . and all weights can't added into 1. I'm not sure how to interpret the API? Could you explain how Snorkel implemented this at high level?
Estimation of label model with probabilistic labelling functions
I'm currently using Snorkel for a sequence labelling task, and I wonder whether Snorkel could be adapted to use labelling functions that provide probabilistic distributions over labels instead of single labels (or abstain). Looking at the label model in label_model.py, it seems…