Results 17 issues of Ayush Baid

We want to try out the averaging algorithm used by the CVPR paper **Revisiting Rotation Averaging: Uncertainties and Robust Losses** and benchmark the results on our datasets. This was also...

We cannot recover the correct poses right now and the unit test requires very high tolerances to pass.

bug

The current style of storing the models as private variables of a class in `__init__`, and reuse them in functions. This leads to trouble as the scheduler gets locked up...

objective
longterm

We use pip to install Argoverse in our conda environment. The Argoverse pins the numpy version to an older version ([ref](https://github.com/argoai/argoverse-api/blob/master/setup.py)). This leads to uninstall of the conda installed numpy,...