* reduce global state
* use a decorator here too
* use that one
* use base imports
* ipchandler in setup
* move to common dir
* move to helpers
* wip
* fix the decorator
* Added modeld.py (WIP)
* No more VisionIpcBufExtra
* Started work on cython bindings for runmodel
* Got ONNXModel cython bindings mostly working, added ModelFrame bindings
* Got modeld main loop running without model eval
* Move everything into ModelState
* Doesn't crash!
* Moved ModelState into modeld.py
* Added driving_pyx
* Added cython bindings for message generation
* Moved CLContext definition to visionipc.pxd
* *facepalm*
* Move cl_pyx into commonmodel_pyx
* Split out ONNXModel into a subclass of RunModel
* Added snpemodel/thneedmodel bindings
* Removed modeld.cc
* Fixed scons for macOS
* Fixed sconscript
* Added flag for thneedmodel
* paths are now relative to openpilot root dir
* Set cl kernel paths in SConscript
* Set LD_PRELOAD=libthneed.so to fix ioctl interception
* Run from root dir
* A few more fixes
* A few more minor fixes
* Use C update_calibration for now to exactly match refs
* Add nav_instructions input
* Link driving_pyx.pyx with transformations
* Checked python FirstOrderFilter against C++ FirstOrderFilter
* Set process name to fix test_onroad
* Revert changes to onnxmodel.cc
* Fixed bad onnx_runner.py path in onnxmodel.cc
* Import all constants from driving.h
* logging -> cloudlog
* pylint import-error suppressions no longer needed?
* Loop in SConscript
* Added parens
* Bump modeld cpu usage in test_onroad
* Get rid of use_nav
* use config_realtime_process
* error message from ioctl sniffer was messing up pyenv
* cast distance_idx to int
* Removed cloudlog.infos in model.run
* Fixed rebase conflicts
* Clean up driving.pxd/pyx
* Fixed linter error
* use scripts
* remove common
* do login in the workflow
* separate step
* cleanup
* one script is cleaner
* keep it consistent
* review suggestions
* move in ci too
* PR cleanup
* use_laika
* Update test
* fix spacing
* fix proc replay
* small fix
* Try correct std
* cast correctly
* experimental
* simpler
* update reset thresholds
* improve
* Ready for live running
* outlier reject
* update laika
* add fine speeds
* more experiments
* fixes
* all in laika
* more small tweaks
* outlier rejection not needed
* fix conflict
* no more gpslocation
* update test
* refactor to laika
* runs again
* Update ref
* build docker image with scons cache
* use setup
* scons cache
* added the dockerfile
* add the shell
* docs use cache
* dont pull lfs for docs
* simulator too
* dont pull the cache image
* fix inputs
* inputs
* booleans aren't supported
* bool not supported
* bool isnt supported
* fix
* readonly cache
* fix sim cache too
* respect readonly
* gitignore and move to correct folder
* also copy old
* move to tools/ci
* move to selfdrive/test
* just make the cache writable, it's discarded anyway
* minimize diff in this pr
* calculate rate
* added adeeb's comment
* move to selfdrive/car
* that makes way more sense
* much simpler
* fix rate
* preglobal doesn't have a counter
* update ref
* merge
* update ref after merging
* radard: tie radard frequency to modelV2
Accumulate parsed messages until state is updated (toyota)
Same for honda
Rename rr to something more descriptive
* Change _update method name, since signature has changed
* Update ref commit
* Some renames
* Check for number of cans
* enable flake8-builtins
* replace any with contains
* fix typo in pack
* fix type
* format is from the parent module, has to be enabled
* item_id
* fix item_id
* disable for id since that's what the remote server returns
* Added all maneuvers to navInstruction message
* Added nav instruction logic to modeld
* New model: fcee01c1-96bb-414f-b00d-e4a994a00922/700
* Fixed bug in navd
* Added sharp/slight modifiers
* Updated refs
* Add ProcessOutputProxy
* Move launcher to its own field
* Move ProcessOutputCapture to its own file
* Return itself from __enter__ of OpenpilotPrefix
* Integrate ProcessOutputCapture into process_replay
* Add note about capture_output_store to README
* ipykernel import is optional
* Disable type checking for link_with_current_proc
* Remove assertion
* Decode outputs to utf-8
* read(self): return empty buf if its none
* Fix type annotations
* Replace fifo with regular file, to avoid hitting fifo size limit
Justin Newberry
Kacper Rączy
Harald Schäfer
Adeeb Shihadeh
Mitchell Goff
Shane Smiskol
Jason Young