github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-24 22:38:16 -05:00
Code Issues Packages Projects Releases Wiki Activity
1,215 Commits 205 Branches 14 Tags
v0.4.0
Commit Graph

50 Commits

Author SHA1 Message Date
George Hotz
e6b65f8e01 fix graph in openpilot/compile.py 2022-10-28 08:55:34 -07:00
George Hotz
ef62db3186 cleanups, remove E701 2022-10-28 08:28:56 -07:00
George Hotz
b65b70812a Exec AST (#404) 
* working exec ast

* exec_ast is staticmethod

* GenericExecAST

* fold that sometimes

* ExplicitExecAST

* exec_ast for GPU

* gpu working

* get_lazyop_shape

* now gpubuffer is ExplicitExecAST

* dedup

* add a type

* RESHAPE in opencl code

* fix linter

* that too for linter

* cleanups

* remove dead code

* GenericShape is less lines

* add ALLOWED_KERNEL_COUNT to tests

* fix mypy

* that's gotta be recursive

* fix opencl shape processing

* remove unneeded lambda
 2022-10-28 08:27:03 -07:00
George Hotz
6a8fb53304 move ops.py into lazy.py (#402) 
* move ops.py into lazy.py

* fix graph and linter

* ugh, didn't add
 2022-10-25 13:58:03 -07:00
George Hotz
3b9b7eda48 remove run_thneed dead code 2022-10-20 17:24:18 -07:00
George Hotz
1bec4651b3 fix nonstatic weights 2022-10-20 17:04:14 -07:00
George Hotz
50c95c7d9a add assert to catch issue in attention 2022-10-20 15:13:00 -07:00
George Hotz
26c78ccf7d remove useless buffer 2022-10-20 14:07:28 -07:00
George Hotz
a18c1f3178 zero out the inputs 2022-10-20 13:46:52 -07:00
George Hotz
61ee428e4c rerun 2022-10-20 13:29:14 -07:00
George Hotz
5dae64b7b0 read input shapes and break down the layers 2022-10-20 13:11:24 -07:00
George Hotz
e00601faea fix thneed self test 2022-10-20 12:55:02 -07:00
George Hotz
ace8db29f8 ReduceSum 2022-10-20 12:48:14 -07:00
George Hotz
c400ee0beb refactoring thneed (#400) 
* refactoring thneed

* continue

* minor update

* looks like it's working

* big refactor

* confirm thneed got the right output

* code is there but it's broken

* works now

* always OPTWG, input -> dat

* fix type issue
 2022-10-20 12:35:59 -07:00
YassineYousfi
ae0f9b17df openpilot: new models and onnx ops (#401) 
* ngrl stuff

* fngrl

* fix typo in compile script

* workflow dispatch

* new models in tests

* dont need to up this threshold

Co-authored-by: HaraldSchafer <harald.the.engineer@gmail.com>
 2022-10-20 11:49:19 -07:00
George Hotz
d6f499fd69 improve opencl, why is it OOMing 2022-09-05 20:14:31 -07:00
George Hotz
2e9b7637b3 don't save input buffers 2022-08-31 15:37:38 -07:00
George Hotz
a3fc64a585 fix batchnorm folding in openpilot compile 2022-08-31 13:04:49 -07:00
Comma Device
a734df98fa TEST_ENET for openpilot compiler 2022-08-31 13:23:36 -04:00
George Hotz
d919ac32af fix wrong size input 2022-08-31 09:07:34 -07:00
George Hotz
040640a580 fix cl import error 2022-08-31 08:43:44 -07:00
George Hotz
33ac355bcd still broken 2022-08-29 19:08:07 -07:00
George Hotz
5efab7cf1d add reciprocal 2022-08-29 18:00:24 -07:00
George Hotz
880707f2d2 no torch test if no torch 2022-08-29 15:29:19 -07:00
George Hotz
5eba228844 print inputs 2022-08-29 08:56:04 -07:00
George Hotz
dd587d26e3 oops, compare with abs 2022-08-28 11:23:21 -07:00
George Hotz
dc7af8c3ac thneed run float32 2022-08-28 11:03:35 -07:00
Comma Device
f0d11f29c7 float32 in image desc 2022-08-28 08:47:43 -07:00
George Hotz
11626053b0 run_thneed with test 2022-08-22 09:45:46 -07:00
George Hotz
e7a4cd91ba fix cpu thneed running 2022-08-21 12:11:07 -07:00
George Hotz
a8734df030 add openpilot tests to tinygrad 2022-08-21 12:03:37 -07:00
Comma Device
85453288d7 run_onnx_torch 2022-08-18 08:30:12 -07:00
Comma Device
1f23517d92 fixup run thneed 2022-08-18 08:22:53 -07:00
Comma Device
6da956b9fa that should be right 2022-07-19 19:47:37 -07:00
Comma Device
f4ed837f2f float16 fixups 2022-07-19 19:44:40 -07:00
Comma Device
aa00a3948e needs_load in image correct 2022-07-19 19:25:47 -07:00
Comma Device
314d70ff17 zero out the buffer 2022-07-19 19:17:47 -07:00
Comma Device
b8a67905e5 save weights 2022-07-19 19:14:14 -07:00
Comma Device
2d402d1135 buffer_id is 8 bytes 2022-07-18 20:27:45 -07:00
Comma Device
577c23731e outputs with size 2022-07-18 20:21:33 -07:00
Comma Device
29581b5c85 inputs and outputs 2022-07-18 20:17:26 -07:00
Comma Device
ae30641b0d fix row pitch 2022-07-18 19:48:19 -07:00
Comma Device
02f23e526c output file to disk 2022-07-18 19:23:22 -07:00
George Hotz
bcf422dfdd Device2 (#358) 
* option for matmul

* fixups

* fast like a nascar

* running

* thneed runner

* no buffer id makes no backing buffer

* move constant folding to the top

* runs on mac

* folded biases

* was v slow

* maybe just that

* elu touchup

* speed and float32

Co-authored-by: Comma Device <device@comma.ai>
 2022-07-16 07:26:19 -07:00
George Hotz
d651caa864 fixup openpilot/compile.py 2022-07-11 13:59:09 -07:00
George Hotz
d8e7f1f8bc opencl type ignore 2022-07-08 10:33:55 -07:00
George Hotz
df7976248b be lazy with the gpubuffer copies for host for constant folding 2022-07-03 23:04:14 -07:00
George Hotz
18d74c01b1 float4 opt 2022-06-21 21:27:51 -07:00
George Hotz
ff3d5fe962 debugging while we compile 2022-06-21 21:12:04 -07:00
George Hotz
b12985b013 openpilot compiler 2022-06-21 20:31:18 -07:00