I built a BNN in TensorFlow and I am trying to make a prediction just using the green dots(see image below), however, my uncertainty bounds do not grow/expand where there is no data. I have attached the code I used to define and build my BNN and I have tried to debug why my uncertainty bounds are not expanding and narrowing at the correct points. Any help would be appreciated.
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Why aleatoric uncertainty converge to log(2) | 0 | 1026 | August 10, 2022 | |
| If probabilistic layer gives a distribution as output and sampling from them using model.predict gives non deterministic values then how can we sure that this is the exact value | 1 | 339 | September 11, 2024 | |
| How to correctly use DenseFlipout with TensorFlow probability | 1 | 1392 | November 3, 2022 | |
| Custom posterior in DenseFlipout layer | 1 | 759 | October 7, 2024 | |
| Help understanding log_prob and defining the right loss function with tensorflow probability | 0 | 1487 | April 25, 2022 |