George Hotz
cfdf803b52
fix llvm vectorization by add analysis passes from the target machine
2022-10-30 15:28:36 -07:00
George Hotz
2f602a92ff
seperate STRIDED and EXPAND
2022-10-30 13:23:58 -07:00
George Hotz
544cb0a069
oops, remove while(1)
2022-10-29 14:05:13 -07:00
George Hotz
4b6097f81d
more amx notes
2022-10-29 14:04:10 -07:00
George Hotz
fdb43fe553
gemm is 1.7 TFLOPS on a single M1 core
2022-10-29 13:42:33 -07:00
George Hotz
52bfbc31be
vectorization
2022-10-29 12:47:52 -07:00
George Hotz
e473d35f90
llvm doesn't vectorize
2022-10-29 11:59:48 -07:00
George Hotz
86eb06eb76
accurate flop estimation
2022-10-28 19:13:20 -07:00
George Hotz
7909786dbf
one more opt test
2022-10-28 18:37:53 -07:00
George Hotz
dd543fbc7a
MovementOps is unused
2022-10-28 18:26:08 -07:00
George Hotz
71b336503f
no RESHAPEs in the AST
2022-10-28 18:25:30 -07:00
George Hotz
294ab9e2f8
more test opt
2022-10-28 18:04:12 -07:00
George Hotz
f885ceb695
test speed w/o bias
2022-10-28 11:22:15 -07:00
George Hotz
3735e26492
very minor
2022-10-28 09:39:30 -07:00
George Hotz
c0050fab8f
clean up movement_op in cpu and torch
2022-10-28 09:29:12 -07:00
George Hotz
df31dde174
hasattr and DeviceBuffer type fixups
2022-10-28 09:05:45 -07:00
George Hotz
e6b65f8e01
fix graph in openpilot/compile.py
2022-10-28 08:55:34 -07:00
George Hotz
1013540370
fix flake8
2022-10-28 08:52:53 -07:00
George Hotz
804b2dd001
move into graph.py
2022-10-28 08:50:11 -07:00
George Hotz
8517b69bfb
lazy cleanups
2022-10-28 08:43:43 -07:00
George Hotz
d02f8f9bc0
can we lose the lines with E701 still there?
2022-10-28 08:36:03 -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
6a15fd3844
LLVM Backend take 2 ( #403 )
...
* take 2 llvm
* get_lazybuffers -> get_buffers
* llvm tests pass
* fix type issues and refactor LLVM
2022-10-26 20:32:31 -07:00
George Hotz
10921a60c4
more imports from llvm branch
2022-10-26 18:02:36 -07:00
George Hotz
463995e64f
relu simpler backward pass
2022-10-26 17:57:32 -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
8e22d5ee67
replace networkx with defaultdict
2022-10-20 19:36:43 -07:00
George Hotz
3b9b7eda48
remove run_thneed dead code
2022-10-20 17:24:18 -07:00
George Hotz
63f9c55156
really dumb bug
2022-10-20 17:07:47 -07:00
George Hotz
1bec4651b3
fix nonstatic weights
2022-10-20 17:04:14 -07:00
George Hotz
59143bbb3b
raise, don't assert
2022-10-20 16:32:34 -07:00
George Hotz
9f8c414589
might fix tests
2022-10-20 16:27:11 -07:00
George Hotz
fd6ba8e7ac
don't recopy backing
2022-10-20 16:06:11 -07:00
George Hotz
62affbd9ce
add CONTIGUOUS loadop
2022-10-20 15:55:19 -07:00
George Hotz
bb288e6938
safe_numpy and warning for broken matmul
2022-10-20 15:40:22 -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
George Hotz
0514594083
fix openpilot test
2022-10-20 11:56:26 -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
Drew Hintz
a4ad1d774a
enable tests in test_ops.py that are disabled but now work. ( #396 )
...
remove custom tolerances that don't appear to be needed.
2022-10-13 09:58:53 -07:00
Drew Hintz
165fb4d631
remove redundant list comprehension from inside all. ( #397 )
...
remove explicit inherit from object.
2022-10-13 09:58:35 -07:00
George Hotz
793edf8900
touchup
2022-10-10 16:13:34 -07:00
George Hotz
d54a45b50d
measure speed vs torch
2022-10-10 16:06:00 -07:00
