429a638740
New map model: 917a2539-5e7a-4374-878a-9b356f0f1922/500 ( #28579 )
...
old-commit-hash: a9cb255017
2 years ago
2fe79435a2
update model replay ref commit
...
old-commit-hash: fc61209f74
2 years ago
4de14c1130
hot coffee model ( #28296 )
...
* 1061b1c7-b944-43e3-a940-b56b64d66f54/700
* bump cereal
* bump cereal
* make mypy happy
* write TODO
* read height from written params
* fix certain_if_calib logic and unit tests
* factor moving_avg_with_linear_decay
* remove whitespace
* update model ref commit
* default value for CI
* update process replay ref commit
* update process replay ref commit
* update process replay ref commit
* bump cereal
old-commit-hash: 7925232a34
2 years ago
3dc3978a10
Test navmodel ( #28272 )
...
* Test navmodel
* removed comment
* Updated model_replay ref
* read tiles from azure bucket
* updated ref commit
old-commit-hash: 2d34083cd8
2 years ago
22c5053bbc
Nicki Minaj Model ( #28218 )
...
* New model: 2ff7490f-8a2f-4c0d-87a8-b0fa3a4a6a71
* Updated model_replay_ref_commit
old-commit-hash: 328b5d9d47
2 years ago
fa6e10b739
new model: trained with augmented turn desire pulses ( #28063 )
...
* df733db3-fad2-4485-abe9-59a6a47f2233/449 85f20e13-dd4d-49a8-ba33-98e6afc161d9/700
* update ref commit
old-commit-hash: f80c1087fc
2 years ago
cde221ba7c
new model: trained on new dataset ( #27637 )
...
* df733db3-fad2-4485-abe9-59a6a47f2233/449 946c0f55-247d-4790-a407-c07e4364b340/700
* fix reducel2 df733db3-fad2-4485-abe9-59a6a47f2233/449 946c0f55-247d-4790-a407-c07e4364b340/700
* update ref commit
old-commit-hash: 26541d9e7a
2 years ago
2fe836f158
tinygrad: bump to 0.5 ( #27505 )
...
* bump tinygrad to 0.5
* update release files
* ref commit + bump tinygrad
* save 220 mW
---------
Co-authored-by: Comma Device <device@comma.ai>
old-commit-hash: 3892c60980
2 years ago
542e915982
new model: improved height estimation ( #27234 )
...
* 07a3a2b2-6dbe-43c3-9bb9-a8023932e054/449 c3d7d18a-6207-42e7-84d9-601ecb3c6f0e/700
* 07a3a2b2-6dbe-43c3-9bb9-a8023932e054/449 db92e7f3-3f4b-4f0f-95ff-76cb6ed6894a/700
* update ref commit
old-commit-hash: 1e49c54ffb
2 years ago
452031fcc7
new model: fixed weight decay logic ( #27068 )
...
* 07a3a2b2-6dbe-43c3-9bb9-a8023932e054/449 cdfa9e3c-2807-49fc-bdc7-a985f8ff1644/700
* update refs
old-commit-hash: 7316cd03fb
2 years ago
94fc9ef358
tinygrad: update to new version ( #27067 )
...
* tinygrad: bump and retain OPENCL behavior
* add graph.py to release, bump tiny
* update files_common and read from it in scons
* bump with typo fix
* switch tinygrad backend to GPU
* update ref commit
Co-authored-by: Comma Device <device@comma.ai>
old-commit-hash: 64ed44e3b5
2 years ago
b3626c8ecd
model_replay: use last N frames of segment so navRoute is valid ( #26790 )
...
* model_replay: use last N frames of segment so navRoute is valid
* Use llk logMonoTime for MapRenderState.locationMonoTime
* Record mapRenderState
* update refs
Co-authored-by: Comma Device <device@comma.ai>
old-commit-hash: 5ab2fc16ab
2 years ago
f315819d47
Navmodel and driving style: update runner ( #26762 )
...
* Navmodel and driving style: update runner
* No driving style yet
* thneedrunner without extra inputs
* nav feature ref
old-commit-hash: 160c9ba1d2
2 years ago
44098cc024
navmodel: check images in replay test ( #26747 )
...
* move position
* log images
* enable test
* update refs
* ignore
Co-authored-by: Comma Device <device@comma.ai>
old-commit-hash: f3a711953a
2 years ago
01db07375c
nav model replay ( #26697 )
...
* nav model replay
* set token
* fix gh actions
* update refs
* fix pc
old-commit-hash: 1da6f6a6e2
2 years ago
30414a7305
Revert "Albert Einstein Model" ( #26576 )
...
Revert "Albert Einstein Model (#26456 )"
This reverts commit 0129c20611bd21fb5b94d19b96d435c7e02041f1.
old-commit-hash: ceb94f1bcd
2 years ago
5bb8ff4a90
Albert Einstein Model ( #26456 )
...
* d0bc247d-9529-4e50-a5dd-95a8a5733874/440 38fb1ce9-18d7-4292-9ffd-678f8483ebf4/700
* update model replay ref commit
old-commit-hash: 34c80a6fbd
2 years ago
f86b44d516
E2e long model: calibrate model speed to wheel speed ( #26395 )
...
* calibrate!
* Fix test
* Fix proc replay
* check len
* get v_ego from model 8501d20-bb59-4193-aa82-82b2737dedd6/449 609d90f3-65e6-4617-a60c-d6d99eead408/700
* bump cereal
* initialize v_model_error
* typo
* better names
* cleanup
* bump cereal
* update model replay ref commit
* bump to cereal master
Co-authored-by: Yassine Yousfi <yyousfi1@binghamton.edu>
old-commit-hash: f63f0de80a
3 years ago
eafe85c1eb
Model trained with headlight augmentation ( #26284 )
...
* d8501d20-bb59-4193-aa82-82b2737dedd6/449 ddfe2a22-fb0e-4ff0-a97f-0157b95eb44d/700
* update ref
old-commit-hash: c3388d216c
3 years ago
4b0d7997ac
Model trained with more long noise ( #26241 )
...
* d8501d20-bb59-4193-aa82-82b2737dedd6/449 d37d320d-7d78-479a-9be2-d58c75284307/700
* Update model ref
old-commit-hash: 2b7290a142
3 years ago
bcb1ba7bfe
Wide transform model ( #26165 )
...
* d8501d20-bb59-4193-aa82-82b2737dedd6/449 f78009d1-2184-4136-9fbc-1c5de55d460d/700
* Change model outs
* Add cereal
* Publish transform
* d8501d20-bb59-4193-aa82-82b2737dedd6/449 f78009d1-2184-4136-9fbc-1c5de55d460d/700
* Bump cereal to master
* Bump cereal to master
* Update model ref
old-commit-hash: d25ab3e366
3 years ago
426286f24f
DM: add use of e2e preds ( #26078 )
...
* try ml
* de56
* j914ef75a
* jd1124586
* jd1124586
* d112
* oops
* set
* update ref
* use offset
* bump DM power usage
* new ref
old-commit-hash: d522492ba0
3 years ago
f428c9de30
Refactor model: no klblock ( #26035 )
...
* ff138dc0-d097-4818-b40e-dba5ba89d5d6/449 13274b7d-b546-4b91-a587-33b4af7dec6a/700
* b1bb39be-c6ce-4744-8e63-92969fda6bfc/449 f3ebfba1-f686-448f-be9b-b4d5010be91c/700
* model ref
Co-authored-by: Yassine Yousfi <yyousfi1@binghamton.edu>
old-commit-hash: b654ebdd25
3 years ago
523ad86756
faster rocket launcher model ( #26009 )
...
* cache tokens 1456d261-d232-4654-8885-4d9fde883894/440 e63ab895-2222-4abd-a9a5-af86bb70e260/700
* udpate ref commit
* bump tinygrad to master
old-commit-hash: 6c5693e965
3 years ago
a8ec6f4cfd
Rocket Launcher Model ( #25963 )
...
* 1456d261-d232-4654-8885-4d9fde883894/440 6b7d7cec-ead8-40f3-86cc-86d52c9b03fe/300
* compute only 9 tokens: 1456d261-d232-4654-8885-4d9fde883894/440 6b7d7cec-ead8-40f3-86cc-86d52c9b03fe/300
* tinygrad: cleanup gather
* 1456d261-d232-4654-8885-4d9fde883894/440 6b7d7cec-ead8-40f3-86cc-86d52c9b03fe/700
* empty commit for tests
* bump tinygrad
* dont use tinygrad matmul for now
* bump tinygrad
* 1456d261-d232-4654-8885-4d9fde883894/440 e63ab895-2222-4abd-a9a5-af86bb70e260/700
* float16 1456d261-d232-4654-8885-4d9fde883894/440 e63ab895-2222-4abd-a9a5-af86bb70e260/700
* increase steer rate cost
* Revert "increase steer rate cost"
This reverts commit 74ce9ab9be7ef17ecfec931f96851b12f37f2336.
* fork tinygrad
* empty commit for tests
* basics
* Kinda works
* new lat
* new tuning
* Move LATMPCN so scons compiles
* Update long weights
* Add tinygrad optim
* Update model ref
* update weights
* Update ref
* Try
* Error message for field ignore
* update model regf
* ref commit
* Fix onnx test
Co-authored-by: Yassine Yousfi <yyousfi1@binghamton.edu>
old-commit-hash: cb0b7375b7
3 years ago
4b8547c7a4
Support e2e long in longitudinal planner ( #25636 )
...
* refactor
* Add planer modes to support offline, acc, and blended
* add acceleration
* Fix index
* Update model ref
* Read in model outputs
* Add model msg
* Add e2e logic
* Add source
old-commit-hash: e1b7a37a1f
3 years ago
29cd51b876
modeld: Move from SNPE to tinygrad ( #25207 )
...
* compiling, won't work yet
* running with inputs and outputs
* there's some magic chance this works
* no more dlc, include onnx
* yolo tests plz
* bump tinygrad
* files_common + delete dlc
* tinygrad_repo -> tinygrad
* pre commit config
* llops needed
* extra in files_common
* bump tinygrad
* fix indent
* tinygrad/nn/__init__
* tinygrad_repo
* bump tinygrad repo
* bump tinygrad
* bump with native_exp, match maybe
* native_explog is argument
* pyopencl no cache
* 5% chance this matches
* work in float32?
* bump tinygrad
* fix build
* no __init__
* fix recip
* dumb hack
* adding thneed PC support
* fix pc segfault
* pc thneed is working
* to_image
* prints stuff with debug=2
* it sort of works
* copy host ptr is simpler
* bug fix
* build on c3
* this correct?
* reenable float16
* fix private, fixup copy_inputs internal
* bump tinygrad and update ref commit
* fix OPTWG on PC
* maybe fix non determinism
* revert model replay ref commit
* comments, init zeroed out buffers
* upd ref commit
* bump tinygrad to fix initial image
* try this ref
Co-authored-by: Comma Device <device@comma.ai>
old-commit-hash: 40d6f4b65c
3 years ago
8413d5ddf1
Nuclear Grade Model: less memory, more accuracy ( #25524 )
...
* c9d10c64-bea4-41ec-8ca3-d8c886fda172/440 6d1c8a6b-4070-4780-80f1-6f08f234275e/900
* update ref
old-commit-hash: 2f46fe5d85
3 years ago
48522500ea
Falcon Punch Model: turn cutting improvements ( #25413 )
...
* simplified change to mpc dynamics
* add jerk pts
* increase jerk cost
* increase jerk pts multipler to master value
* Add final commit
* 1456d261-d232-4654-8885-4d9fde883894/440 ac1a6744-85b0-4ec6-8ba7-608d0717b8f1/750
* some copies are useful
* update model replay ref
* less frames in model replay onnx cpu
* 1456d261-d232-4654-8885-4d9fde883894/440 264b67f5-3f52-4b58-b11f-58dd8aaf08bf/950
* 1456d261-d232-4654-8885-4d9fde883894/440 236fc556-fba3-4255-8ccf-684b22637160/950
* c9d10c64-bea4-41ec-8ca3-d8c886fda172/440 26d73dd2-862a-44ae-bbdd-32cc4f397ad7/900
* Fix couple tests
* Update ref
* Unused for now
* Add lateral factor comment
* Unused variable
Co-authored-by: nuwandavek <vivekaithal44@gmail.com>
Co-authored-by: Bruce Wayne <yassine@comma.ai>
Co-authored-by: Yassine Yousfi <yyousfi1@binghamton.edu>
Co-authored-by: Bruce Wayne <batman@gpu06.internal>
old-commit-hash: 041458f632
3 years ago
6709805020
Model file too big: Revert "New model: minor fixes" ( #25390 )
...
Revert "New model: minor fixes (#25368 )"
This reverts commit 4e69dfab07507510eb52f532e7038b01f83471b7.
old-commit-hash: 3afe7464df
3 years ago
f8a05617a3
New model: minor fixes ( #25368 )
...
* 1456d261-d232-4654-8885-4d9fde883894/440 ac1a6744-85b0-4ec6-8ba7-608d0717b8f1/750
* some copies are useful
* update model replay ref
* less frames in model replay onnx cpu
Co-authored-by: Bruce Wayne <yassine@comma.ai>
Co-authored-by: Yassine Yousfi <yyousfi1@binghamton.edu>
old-commit-hash: 55bf938504
3 years ago
f7d1aad21e
Update model replay commit for calib change ( #25357 )
...
* Update model replay commit for calib test
* more reliable baseline
* update docker login
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
old-commit-hash: 9fbbefbc94
3 years ago
a9108f29fa
Revert "Rocket league model ( #24869 )" ( #24936 )
...
* Revert rocket league
* revert ref commit
* New model ref commit
old-commit-hash: 0e0b5c4e24
3 years ago
cd2a98aa11
fullframe DM model ( #24860 )
...
* Revert "put cereal on master"
This reverts commit 4646c132bae7358079c9b2867725f8382906c1e5.
* Revert "Revert fullframe DM model (#24812 )"
This reverts commit 59e8af4c3101785cead69a9880cc03e0a18081e1.
* revert revert cereal
* clip6
* 0.8 is fair
* Fiction compensation should be based on error
* Update refs
* Add deadzone
* not that
* good mg
* ref
* ref
* ee8f
* minor tweak
* ref
* recompile
* ref
* cereal
* match driverstatus
* new ref
* new ref
* pass token through jenkins credentials
* quote
* fix snpe dead weights
* final ref
Co-authored-by: Harald Schafer <harald.the.engineer@gmail.com>
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
old-commit-hash: 1f2f9ea9c9
3 years ago
b0b1aff5cc
Rocket league model ( #24869 )
...
* dd9a502d-c8e2-4831-b365-804b0ae0739d/600 80041070-d276-4fed-bdb9-0075e5442908/420
* no elementwise op
* 9dabf0fe-2e60-44bf-8d3a-d20a74aca072/600 ae746590-0bb5-4a16-80db-15f02d314f03/300 c4663a12-b499-4c9b-90dd-b169e3948cb1/60
* explicit slice
* some copies are useful
* 1456d261-d232-4654-8885-4d9fde883894/440 c06eba55-1931-4e00-9d63-acad00161be0/700 af2eb6ba-1935-4318-aaf8-868db81a4932/425
* 154f663e-d3e9-4020-ad49-0e640588ebbe/399 badb5e69-504f-4544-a99e-ba75ed204b74/800 08330327-7663-4874-af7a-dcbd2c994ba7/800
* set steer rate cost to 1.0
* smaller temporal size
* Update model reg
* update model ref again
* This did upload somehow
* Update steer rate cost
Co-authored-by: Yassine Yousfi <yyousfi1@binghamton.edu>
old-commit-hash: 9283040d84
3 years ago
07c859d384
Revert "Update model ref"
...
This reverts commit de5133d5378448a18e1e829ac2136cba07f4e346.
old-commit-hash: c851cf7379
3 years ago
1d4ba4670b
Update model ref
...
old-commit-hash: efc8aa05b4
3 years ago
33bed6d13b
remove weights fixup with new SNPE ( #24254 )
...
* remove weights fixup with new SNPE
* Update ref
Co-authored-by: Comma Device <device@comma.ai>
Co-authored-by: Harald Schafer <harald.the.engineer@gmail.com>
old-commit-hash: 7a6f57a28e
3 years ago
c73d4e5fc6
Revert fullframe DM model ( #24812 )
...
* Revert "fullframe DM: flip RHD yaw to use matching thresholds"
This reverts commit ce7daabc8847d18ba46e5d1879f5a6958d04ccc7.
* Revert "fullframe DM model (#24762 )"
This reverts commit 817be81fb19004f4873881f6b29dcdfffbe7e3a8.
* revert cereal
old-commit-hash: c646eeee0a
3 years ago
02f0f39b52
fullframe DM model ( #24762 )
...
* get log
* simplify two nonsense
* not needed
* libyuv is a joke
* clean up
* try small
* fast but not bad
* working
* clean up driverview
* simplified
* thats mirrored
* smol
* tweak
* ref is screen
* w/ ee
* update camera model
* no if TICI
* start
* update pose thresh
* less cpu more dsp
* new libyuv
* new snpe
* add files
* test
* should be fast
* update out len
* trigger test
* use master snpe
* add cereal
* update cereal
* refactor parsing
* missing ;
* get
* wrong type
* test model
* use driver data
* 10829278-72fe-4283-a118-2cef959ce174/1550
* no pf
* adapt driverview
* ;
* rhd learner
* update libyuv buildi x64
* ad4337ea
* remove blink slack
* test
* no
* use toggle
* b16
* fix for nv12
* 5b02cff5 both
* update test
* update cereal
* update cereal
* update cereal
* v2 packets
* revert libyuv
* no /
* update snpemodel
* ;
* memcpy
* fix test
* use toggle in driverview
* update power
* update replay
* Revert "update replay"
This reverts commit 1d0979ca59d6c07a6b15
3 years ago
55085a12db
New model: better laneless ( #24605 )
...
* Fix ping pong
* Fix hugging: e6fd3a62-aa86-46d1-a414-19137ed1babc/800
* Remove lateral toggle
* a6da6265-1952-48dd-986b-2b5996dd2642/950
* Model replay commit
* Revert "Remove lateral toggle"
This reverts commit 69e57eb7e17da48ebdba
3 years ago
3494ec7bbd
Revert "Joint model ( #24521 )" ( #24581 )
...
This reverts commit 9fc732468ee4d4fd135d679a919c748633f06d02.
old-commit-hash: e2e5378dd3
3 years ago
78c2e16a03
Joint model ( #24521 )
...
* update models
* wip lanelines mhp parsing
* typos
* newer models
* 95995a49-db0c-4261-8776-b90780dc2a8c/600
* Use laneful policy to prevent planner aggression
* c8c3ab69-bd2c-4d1c-a7f8-9e8457e4827b/950
* ede8f75a-df3e-41fb-bc56-63e2f33858e5/950
* 7205c5af-7532-4215-ad9b-678789054db7/995
* bfb34fa3-1d72-4987-854f-43d42ee01015/950
* change cost
* 8ae2477a-b452-4601-b04c-d3af66ea5e98/950
* Update heading cost
* 9b991314-94f5-4f0b-872e-58d95266e4d0/333
* 9b991314-94f5-4f0b-872e-58d95266e4d0/950
* ede8f75a-df3e-41fb-bc56-63e2f33858e5/950
* undo lateral planner changes
* rm whitespace
* update ref
* c0acc19d-f2f1-47ae-8838-5382c6c6c7f5/950
* c0acc19d-f2f1-47ae-8838-5382c6c6c7f5/950
* update model refg
* Update model ref
Co-authored-by: Yassine Yousfi <yyousfi1@binghamton.edu>
old-commit-hash: 7996fd37ef
3 years ago
aefc73be1b
Revert "Joint Model ( #24213 )"
...
This reverts commit f6aaafb8d9332c934373fc6fde27abd596b7049f.
old-commit-hash: ef81afc354
3 years ago
4709fb5f61
Joint Model ( #24213 )
...
* update models
* wip lanelines mhp parsing
* typos
* newer models
* 95995a49-db0c-4261-8776-b90780dc2a8c/600
* Use laneful policy to prevent planner aggression
* c8c3ab69-bd2c-4d1c-a7f8-9e8457e4827b/950
* ede8f75a-df3e-41fb-bc56-63e2f33858e5/950
* 7205c5af-7532-4215-ad9b-678789054db7/995
* bfb34fa3-1d72-4987-854f-43d42ee01015/950
* change cost
* 8ae2477a-b452-4601-b04c-d3af66ea5e98/950
* Update heading cost
* 9b991314-94f5-4f0b-872e-58d95266e4d0/333
* 9b991314-94f5-4f0b-872e-58d95266e4d0/950
* ede8f75a-df3e-41fb-bc56-63e2f33858e5/950
* undo lateral planner changes
* rm whitespace
* update ref
Co-authored-by: Harald Schafer <harald.the.engineer@gmail.com>
old-commit-hash: eff08cb17f
3 years ago
eeeec4c2f4
DM: reduce model error on dsp by another 3.5x ( #24159 )
...
* a8236e30
* match
* update model ref
old-commit-hash: 720adbc49f
3 years ago
e8a8e37185
DM: more precision running on DSP + e2e outputs ( #23900 )
...
* update cereal
* run but not use
* log distraction type
* regression scaling
* clean up naming
* add calib buf
* add to header
* fake model
* no calib model
* adjust threshs
* 018a305f
* fix bn
* tweak1
* tweak2
* 0ff2/666
* tweak3
* t4
* t5
* fix out of bound
* skip when replaying old segments
* update ref
* fix onnxmodel
* get calib
* update model replay refs
* up ref
old-commit-hash: de4031c98e
3 years ago
47aec23fb3
modeld: sync based on frame timestamp ( #23901 )
...
* sync based on frame ts
* log extra frame id
* ints only
* update ref
* no rounding errors
old-commit-hash: 492398826f
3 years ago
6b52bc6c13
model: mse err from 0.02-> 0.000056 ( #23891 )
...
* mse err from 0.028070712 -> 5.8073703e-05
* build with weights fixup
* need thneed lib also
* don't break for binaries
* static analysis says i need init
* check the bias
* load_dlc_weights
* nicer scons
* tested scons
* fix static
* pylint issue
* new ref
* a few more asserts
Co-authored-by: Harald Schafer <harald.the.engineer@gmail.com>
old-commit-hash: 8d6f49aecf
3 years ago
fc042cd85a
Fix tici bigmodel ( #23817 )
...
* Fix bigmodel on tici: cam intrinsics arent static
* Give TICI route own name for import
* Make function useable in offline test
* typo
* change tici ecam focal to trained focal
* No shadow variable
* Fixed ref
* Force update this time
* Update model replay ref commit
old-commit-hash: 48154703a7
3 years ago
Mitchell Goff
Yassine
YassineYousfi
George Hotz
Harald Schäfer
Adeeb Shihadeh
ZwX1616
Willem Melching