Results 114 comments of John Lambert

Hi Dmytro, being able to specify a random seed in the python functions as an additional argument would help make unit tests on my end deterministic. Would be great to...

Hi @awethaileslassie, thanks for the heads up. Which log id is this from? Did you run ```bash python waymo_raw_data_to_argoverse.py ``` (can you post your full command here)?

Thanks @awethaileslassie. Unfortunately I don't have time to debug this much this week, but I'm sure it's an easy fix. If you have time, one suggestion would be: take the...

Hi @qiaozhijian, thanks for your note. There are expected to be a few minor errors from human labeling, as more than 10,000 frames are labeled and evaluated. However, let me...

@ayushbaid CI appears to still be failing: ``` =========================== short test summary info ============================ FAILED tests/averaging/rotation/test_shonan.py::TestShonanRotationAveraging::test_circle_all_edges - AssertionError: False is not true ====== 1 failed, 435 passed, 7 skipped, 21...

Gets rid of translation error outliers: Before: ![Screen Shot 2021-09-21 at 2 22 58 PM](https://user-images.githubusercontent.com/16724970/134226423-820f5c00-0321-4ef3-a91a-905e02281602.png) After: ![Screen Shot 2021-09-21 at 2 23 03 PM](https://user-images.githubusercontent.com/16724970/134226407-f048eb32-56b4-4ac3-904e-7fd6e1cb7b7e.png)

> How does this solve #306? 306 notes that the "inlier ratio w.r.t. GT model" was far higher than "inlier ratio w.r.t. est model". We would actually expect the following...