mirror of https://github.com/dragonpilot/dragonpilot.git synced 2026-02-24 02:33:53 +08:00

Files

ZwX1616 8605ba8ab9 pyTorch DM (#19760 )

* new model files

* no more rot

* no rot no diag switch

* should be correct and cache aware

* 405d7aeb

* parse new outputs

* change uncertain policy accordingly

* 1bdbd9ed

* adjust

* 0.5 is fine

* 3d4f9cab

* no face prob no loss

* clean up

* clean up test

* update refs

Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>

2021-01-28 14:02:58 -08:00

__init__.py

selfdrive/test

2020-01-17 11:16:14 -08:00

camera_replay.py

remove old model packet (#19769 )

2021-01-19 17:15:16 -08:00

compare_logs.py

add modelV2 to model replay (#2725 )

2020-12-08 22:01:34 -08:00

inject_model.py

Added model_replay and fixed some bugs in camera_replay (#2679 )

2020-12-08 19:01:59 -08:00

model_ref_commit

Nissan port - dashcam only (#1203 )

2020-03-05 15:40:07 -08:00

model_replay_ref_commit

tici jenkins build (#19505 )

2021-01-23 18:34:18 -08:00

process_replay.py

manager.py broadcast process states (#19880 )

2021-01-26 14:49:55 +01:00

README.md

Process replay for C++ daemons (#2288 )

2020-10-12 11:18:53 +02:00

ref_commit

pyTorch DM (#19760 )

2021-01-28 14:02:58 -08:00

test_processes.py

remove old model packet (#19769 )

2021-01-19 17:15:16 -08:00

update_model.py

add type hints to selfdrive/version.py

2020-09-19 14:55:27 -07:00

update_refs.py

add type hints to selfdrive/version.py

2020-09-19 14:55:27 -07:00

README.md

process replay

Process replay is a regression test designed to identify any changes in the output of a process. This test replays a segment through individual processes and compares the output to a known good replay. Each make is represented in the test with a segment.

If the test fails, make sure that you didn't unintentionally change anything. If there are intentional changes, the reference logs will be updated.

Use test_processes.py to run the test locally.

Currently the following processes are tested:

controlsd
radard
plannerd
calibrationd
ubloxd

Forks

openpilot forks can use this test with their own reference logs

To generate new logs:

./update_refs.py --no-upload

Then, check in the new logs using git-lfs. Make sure to also include the updated ref_commit file.