search

uzh-rpg / deep_uncertainty_estimation

This repository provides the code used to implement the framework to provide deep learning models with total uncertainty estimates as described in "A General Framework for Uncertainty Estimation in Deep Learning" (Loquercio, Segù, Scaramuzza. RA-L 2020).
MIT License
121 stars 22 forks source link

passing the varience to the function compute_log_likelihood() in file eval.py #5

Open MoHassoubah opened 3 years ago

MoHassoubah commented 3 years ago

Hello ,

thank you very much for your effort,

I have a question, in file eval.py the function compute_preds(), returns the variance of the model "model_variance", then the variable "outputs_variance" is passed to the function compute_log_likelihood() as "sigma" yet in the function compute_log_likelihood(), the function torch.distributions.normal.Normal() takes "loc" which is the mean of the distribution and "scale" which is the standard deviation yet you are passing the variance not the std to "scale"??

MoHassoubah commented 3 years ago

@mattiasegu

mattiasegu commented 3 years ago

Hi @MoHassoubah, you are right, it should be the std! I fixed this already in the version of the repo on my github: https://github.com/mattiasegu/uncertainty_estimation_deep_learning