ca38c54333
this should break - Revert "use all samples"
...
This reverts commit 6c0d7943ac
7 months ago
6c0d7943ac
use all samples
7 months ago
e5ba737db0
Revert "Filet o Fish model ( #34637 )"
...
This reverts commit 159b1c9eb4
7 months ago
159b1c9eb4
Filet o Fish model ( #34637 )
...
* 690b01c3 seems ok
* correct temporal
* push
* inplace
* bs
* what thw
* is this wrong
* frames are skipped
* new models
* simplify decimation
* clean up
* clean up modelframe
* need attr
* lint
* 0
---------
Co-authored-by: Comma Device <device@comma.ai>
7 months ago
06f0a8e196
0
7 months ago
a26dc46003
lint
7 months ago
aace234cc8
need attr
7 months ago
cfa9f00994
clean up modelframe
7 months ago
8329c2b75b
clean up
7 months ago
6088732628
simplify decimation
7 months ago
c32c405f65
remove model shims ( #34750 )
7 months ago
0abe348283
gc old stuff in modeld
7 months ago
1af8e4c510
new models
7 months ago
29ca4aa7de
modeld refactor: supercombo -> vision+policy ( #34718 )
...
* should be ok
* update tg
* type
* wrong model
* is no more
* update readme
* add back
---------
Co-authored-by: Comma Device <device@comma.ai>
7 months ago
d41311ca20
Switch PC running to tinygrad with LLVM ( #34660 )
...
* try llvm on cpu
* update lock
* Make sure test fails
* With BEAM=3 its fast
* small fixes
* more fixes
* Fix macos
* do dm too
* rm ort
7 months ago
568cf249f8
frames are skipped
7 months ago
896a490db4
is this wrong
7 months ago
e86dd996da
what thw
7 months ago
463bba4b42
bs
7 months ago
389194c482
inplace
7 months ago
ab46f9c0b6
push
7 months ago
5aff3030d2
correct temporal
7 months ago
9ec69ddb36
modeld: trigger model metadata on script change ( #34623 )
7 months ago
a7fc34b23e
690b01c3 seems ok
7 months ago
f06701ea24
modeld: add model names to metadata ( #34602 )
...
* modeld: add model names to metadata
* lint
* type hint
* oops
* assert
* ok Any
7 months ago
d98a554b17
tici: set core affinity for all realtime processes ( #34574 )
8 months ago
2eb3585dae
Not Too Shabby Model ( #34484 )
...
* b25c27a8-263b-4f3f-b1f8-72ac49b4f9c7/700
* replay
* model_replay: support empty
8 months ago
8bfce0979d
modeld: use temporalPose from model ( #34460 )
...
* modeld: use temporalPose from model
* cleanup
8 months ago
1d86afdf83
carName -> brand ( #34474 )
...
* bump
* brand
* body is lower
8 months ago
02ec9e5bf8
dmonitoringmodeld: clean exit ( #34454 )
...
* nice exit
* correct spacing
* sentry
8 months ago
205c2b63aa
modeld: simplify model message construction by removing redundant variables ( #34354 )
...
Simplify model message construction by removing redundant variables
8 months ago
1068779294
modeld: properly release OpenCL context in `__dealloc__` method ( #34353 )
...
release OpenCL context in __dealloc__
9 months ago
c8264cbf2c
Notre Dame model in tinygrad ( #34324 )
...
* release model: 6f23a03f-486b-4d3e-a314-19d149644c7c/700
* old style model in tinygrad
* fix desire
* tg hack
* 20Hz
* no gas probs
* No gas here
* better indexing
---------
Co-authored-by: Yassine Yousfi <yyousfi1@binghamton.edu>
9 months ago
420d6f0d5e
Quick GC pass heading into 2025 ( #34330 )
...
* first pass
* bye bye snpe
9 months ago
1a7c284445
National Public Radio Model 📻 ( #34259 )
...
* f3a009b7-dcb9-41f3-8917-6fcb3cec37bf/400
* 65f26b40-56c9-4c6c-a3ac-e1788bd52567/400
9 months ago
17ca6389e1
Tinygrad runner ( #34261 )
...
* squash
* dmonitoringmodeld: use cl transform (#34235 )
* needs cleanup
* only if tici
* bump tinygrad
* check width
* base modelframe
* .
* need to be args
* more cleanup
* no _frame in base
* tici only
* its DrivingModelFrame
* .6 is fair
---------
Co-authored-by: Comma Device <device@comma.ai>
* Update tinygrad
* tg upstream
* bump tg
* bump tg
* debug
* attr
* misc cleanup
* whitespace
* remove
* Add TODOs to make python proc for modelrunners
* whitespace
---------
Co-authored-by: ZwX1616 <zwx1616@gmail.com>
Co-authored-by: Comma Device <device@comma.ai>
Co-authored-by: Maxime Desroches <desroches.maxime@gmail.com>
9 months ago
7352e612a2
dmonitoringmodeld: use cl transform without tinygrad ( #34266 )
...
* merge
* why
* self.buf_size
* 0.05 more than with tg due to copy
---------
Co-authored-by: Comma Device <device@comma.ai>
10 months ago
5018cf75ff
North America Model ( #34260 )
...
* 0e1c9c12-0472-4a0c-8963-611ad240ec62/400
* rm outputs
10 months ago
a98210aeec
modeld: ort helpers ( #34258 )
...
* ort helpers
* import from ort helpers
* import that too
* linter
* linter
* linter
10 months ago
f6885dcbec
Revert Tinygrad ( #34243 )
...
* Revert "dmonitoringmodeld: use cl transform (#34235 )"
This reverts commit 684b0b9d4d31606a7d1544f58ff758#34171 )"
This reverts commit 7b5a4fbb03
10 months ago
684b0b9d4d
dmonitoringmodeld: use cl transform ( #34235 )
...
* needs cleanup
* only if tici
* bump tinygrad
* check width
* base modelframe
* .
* need to be args
* more cleanup
* no _frame in base
* tici only
* its DrivingModelFrame
* .6 is fair
---------
Co-authored-by: Comma Device <device@comma.ai>
10 months ago
8743bc4fe2
Null Pointer Model ( #34111 )
...
* e8cb7f27-e448-4c15-90c2-ac440cd5a042/400
* 0078ad07-4d46-4086-820f-23d61c90e07f/400
* 4bd74082-70af-47da-8156-e84ebf4d4812/400
* 2a074022-5c2c-4628-97f9-f54849a936a6/400
* 0660aa81-93c5-41b7-9cc2-dc8816a512cd/400
* Clip curvature to reasonable limits
* Better curvature and speed clips
* typo
* typo
* 31aa62c3-b373-4878-8f2e-5107305de187/400
* 384690ca-9b8a-41fe-9bcd-389b20fc6aa4/400
* ref commit
---------
Co-authored-by: Yassine <yassine.y10@gmail.com>
10 months ago
31606a7d15
load model before calling convert_fp16_to_fp32
10 months ago
7b5a4fbb03
Tinygrad runner ( #34171 )
...
* squash
* bump tg
* bump tg
* debump tinygrad
* bump tinygrad
* bump tg
* Skip init iteration
* fixes
* cleanups
* skip first test sample
* typos
* linter unhappy
* update cpu usage
* OPENCL just zeros for now
* imports
* Try printing
* Runs again, but slower
* unused import
* Allow more buffer with tg and all on gpu
* bump tinygrad
* seems ok
* stricter timings for driving looser for dm
* try llvm
* check nvidia
* More timeout for now
* make test pass
* Revert "try llvm"
This reverts commit ef136e478320101fea262bae3579e558da991902.
* small fixes
* whitespace
* revert test timeout
* No model runners
* Always CPU always fast
* No onnx runtime GPU
* more cores
* cleanup
* Is this faster
* Is this faster
* at least runs
* FP32 is faster than 16
* fix deps
* whitespace
* comment
---------
Co-authored-by: Adeeb Shihadeh <adeebshihadeh@gmail.com>
10 months ago
216ebcaa50
Fix model runtime on PC ( #34210 )
...
exhaustive
10 months ago
57fc4f79d1
Postal Service Model ( #34183 )
...
0d68d465-9938-4327-a015-0cef58bc9f3a/400
10 months ago
7a4169379d
Revert "PlayStation® model ( #34133 )" ( #34170 )
...
This reverts commit 5160bee543
10 months ago
1ec2c56b4e
gc old DM init param
10 months ago
5160bee543
PlayStation® model ( #34133 )
...
* e2a90efd-8986-4c97-8f55-22725cdee8b9/400
* bea47b6f-c2f9-460f-9658-73f202756294/400
* 7e8efd49-f1a4-4355-912e-610633e0e7f9/390
* 7e8efd49-f1a4-4355-912e-610633e0e7f9/400
---------
Co-authored-by: Bruce Wayne <harald.the.engineer@gmail.com>
10 months ago
81252f549c
Alabama model ( #34108 )
...
e22d5e0c-1441-4953-b878-8f0f36e527c6/400
10 months ago
ZwX1616
Adeeb Shihadeh
Harald Schäfer
Comma Device
YassineYousfi
Shane Smiskol
Dean Lee
mitchellgoffpc
Maxime Desroches