Benchmarking uncertainty paper

[deebuls]

https://github.com/fregu856/evaluating_bdl

It evaluates mc-dropout and ensemble using some synthetic images .

Just read and write a small paragraph on what it evaluates, what dataset used, how was it evaluated ?

[deebuls]

https://datasets-benchmarks-proceedings.neurips.cc/paper/2021/file/ac1dd209cbcc5e5d1c6e28598e8cbbe8-Paper-round2.pdf

This is the benchmark dataset for Bayessian Deep learning