github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-27 07:48:07 -05:00
Code Issues Packages Projects Releases Wiki Activity
10,417 Commits 207 Branches 14 Tags
reprocess_node_experiments
Commit Graph

10417 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
qazal
e6d20c47e3 simpler becomes_map update [pr] (#9201) 
* simpler becomes_map update

* err, no metadata for device

* simpler tensor metadata mapping + tests [pr]

* remove kernel metadata

* don't map nones

* pruning

* linter
 2025-02-22 20:50:58 +01:00
qazal
4578c3e8fd simpler tensor metadata mapping + tests [pr] (#9203) 
* simpler tensor metadata mapping + tests [pr]

* remove kernel metadata

* don't map nones
 2025-02-22 20:18:46 +01:00
qazal
b711c6343a no early return + allow childless const/bind/var in kernel graph [pr] (#9202) 2025-02-22 19:28:22 +01:00
George Hotz
97bc723538 torch backend works for ResNet-18 (#9200) 
* torch backend progress, a few more functions

* resnet works

* pillow

* tv
 2025-02-22 22:16:23 +08:00
George Hotz
f92820d30d torch backend tests (#9198) 
* torch backend tests

* pythonpath

* install ninja
 2025-02-22 16:01:49 +08:00
George Hotz
4e6665bda5 different way to write torch backend (#9197) 
* different way to write torch backend

* both backends

* more work

* simpler code

* more work

* test both

* imply unwrap/wrap

* FORWARD_ONLY=1 TINY_BACKEND=1 python3 test/test_ops.py TestOps.test_add works

* ready to start making test_ops work in torch backend

* backward pass, TINY_BACKEND=1 python3 test/test_ops.py TestOps.test_add works

* FORWARD_ONLY=1 TINY_BACKEND=1 python3 test/test_ops.py TestOps.test_simple_conv2d works

* matmul backward is broken with as_strided
 2025-02-22 14:42:26 +08:00
nimlgen
041b6d5678 am: load fw in batches (#9185) 
* am: load fw in batches

* am: 1mb less fw copies

* mypy

* list
 2025-02-21 23:21:31 +03:00
qazal
1db4341e9f move viz graph to lib/graph [pr] (#9196) 
* move viz graph to lib/graph [pr]

* add package

* share with program
 2025-02-21 21:04:07 +01:00
geohotstan
6587c7879b simple fixes to onnx (#9195) 
* uncontroversial changes

* cleaner _prepare_quantize
 2025-02-21 13:10:06 -05:00
Simon R
2318d7ac51 Add missing tinygrad.runtime.autogen.am to packages (#9194) 2025-02-21 15:38:24 +02:00
qazal
8bb80b6e5e reorder AST matchers + comments [pr] (#9193) 2025-02-21 14:31:15 +01:00
qazal
2eab8021fb remove inputs+outputs attributes from ScheduleItem [pr] (#9192) 
* remove inputs/outputs from ScheduleItem

* fix test_linearizer

* fix test_conv_shapetracker

* fix test_schedule + lint

* test_image_dtype + multitensor + search
 2025-02-21 13:48:11 +01:00
George Hotz
e87be0131e torch backend start (#9191) 
* start torch backend

* progress

* ugh, you need cpp crap

* 1+1 works

* 1+1 works

* becoming a real backend

* ready to merge?
 2025-02-21 16:57:28 +08:00
George Hotz
d3a21cced2 hotfix: bump version to 0.10.2 v0.10.2 2025-02-21 10:43:49 +08:00
chenyu
2e7c2780a9 CLANG -> CPU (#9189) 2025-02-20 18:03:09 -05:00
nimlgen
f986e12f91 metal: choose compile spec based on macos (#9188) 
* metal: choose compile spec based on macos

* correction
 2025-02-21 00:43:39 +03:00
chenyu
3e22747799 run unit test on windows ci (#9187) 
* factor out testing_minimal in setup.py [pr]

* testing_unit + windows
 2025-02-20 14:40:41 -05:00
chenyu
287de4ecc6 use torch in test_gradient (#9186) 
used torch.autograd.grad, but not sure if it can be a template like jax
 2025-02-20 12:26:11 -05:00
qazal
574a905291 Fix running VIZ=1 after package installation + test (#9183) 
* test running viz from pip install

* add pkg

* do 10 connection attempts

* include assets in package_data

* quiet curl

* better print
 2025-02-20 15:02:00 +01:00
chenyu
1692087db5 _one_hot_along_dim input needs to be int (#9179) 
* _one_hot_along_dim input needs to be int

indexing and onehot compare with arange, and non-int dtype is likely a bug
 2025-02-20 09:00:43 -05:00
George Hotz
bf36967883 cuda hooking (#9180) 
* cuda hooking

* progress

* more hook cuda

* fix params

* compile + cuMemHostAlloc hook

* work

* revert that
 2025-02-20 19:20:01 +08:00
chenyu
3b37cc898b add bert tiny config (#9177) 
set with BERT_SIZE=tiny. easier to study embedding and fusion
 2025-02-19 14:57:03 -05:00
qazal
5662c898f1 correctly step through bottom_up_rewrites in viz [pr] (#9176) 2025-02-19 19:20:57 +01:00
peppingdore
b1ddb2a1a6 fix win32 CPUProgram missing cache flush (#9171) 
* win32: fix missing inst cache flush, rename ptr->self.mem for consistency with posix code

* fix types, remove assert

* fix memory leak

* rm whitespace
 2025-02-19 21:38:51 +08:00
qazal
1bb9d78c7a hotfix: add output buffer back to kernel parents + comment [pr] (#9174) 2025-02-19 14:22:01 +01:00
chenyu
975c318dbc bert use int32 for input ids (#9173) 
original data was int32 for these. float might have caused precision issues
 2025-02-19 08:17:27 -05:00
qazal
e4a8bf28ea scheduler cleanups + better cycle assert [pr] (#9172) 
* scheduler cleanups + better cycle assert [pr]

* type_verify after assign fixup

* don't need base

* always realize sink parents
 2025-02-19 13:30:58 +01:00
qazal
cf315d544b rename can_pad arg to cache [pr] (#9170) 2025-02-19 12:24:59 +01:00
qazal
2fc8bf115d remove support for VIEW with two sources in ops [pr] (#9168) 
* only 1 src views can exist [pr]

* views can still exist without a base, this is a separate project
 2025-02-19 11:10:18 +01:00
Ahmed Harmouche
a2afa523a0 Only add enable f16 directive if ShaderF16 is supported (#9163) 
* F16 in check in wgsl renderer

* Default in renderer to fix pickle

* Refactor f16 check
 2025-02-19 17:20:03 +08:00
Ahmed Harmouche
0f94b98646 Force WebGPU backend type [pr] (#9164) 
* Force webgpu backend type

* Mypy fix

* Rename to WEBGPU_BACKEND

* Add it to env_vars docs

* Remove link
 2025-02-19 17:19:39 +08:00
qazal
4bc708a9b0 do not create buffers we never realize in scheduler (#9165) 
* work

* delete

* fix

* works

* FUSE_CONV_BW

* FUSE_ARANGE

* becomes_map

* fix assign p1

* fix assign (diamond) - 2

* fix test_assign_double_diamond_reduce

* fix subbuffer

* faster rewrite

* fix simple_pads

* start metadata work

* do some diff cleanups

* make things that can't be images not images

* openpilot fix

* fix linter

* diff

* minimal diff

* more work on the diff

* metadata
 2025-02-19 10:11:47 +01:00
George Hotz
1c4e9bc363 image fixup tensor map [pr] (#8611) 
Co-authored-by: qazal <qazal.software@gmail.com>
 2025-02-19 10:11:06 +02:00
qazal
2a5fe3e700 whitespace changes from the map_tensors branch [pr] (#9167) 2025-02-19 09:52:59 +02:00
qazal
a773ff73e3 match image cast folding on the cast itself [pr] (#9166) 2025-02-19 09:31:34 +02:00
qazal
9a20063837 create subbuffer immediately before constructing ScheduleItem [pr] (#9162) 2025-02-18 21:07:52 +01:00
qazal
1c92534bff hotfix: viz should show if there's a rewrite [pr] (#9161) 2025-02-18 19:11:03 +01:00
George Hotz
a330f3338c save applied opts in ProgramSpec [pr] (#9150) 2025-02-19 00:40:03 +08:00
chenyu
ff05bff221 put bert data shard inside jit (#9160) 
python time 45ms -> 9ms, it was spending time to schedule the shard

also init bert data on CLANG since it's from numpy, so we don't create the tensor on default device then shard into GPUS
 2025-02-18 10:36:54 -05:00
qazal
679291e26a assert only base maps to buffer [pr] (#9159) 2025-02-18 15:46:47 +01:00
qazal
4f592eeea6 hotfix: remove extra matcher for copy/buffer_view [pr] (#9157) 2025-02-18 13:21:24 +01:00
George Hotz
ff9b985d9f hotfix: View Base AST 2025-02-18 18:48:34 +08:00
George Hotz
30f470eaa3 UNIQUE UOp for buffer instead of arg (#9156) 
* UNIQUE UOp for buffer instead of arg

* factor out buffer spec
 2025-02-18 16:59:59 +08:00
qazal
38f5ea2132 increment writable buffers refcount from the kernel graph [pr] (#9153) 2025-02-18 10:20:02 +02:00
George Hotz
ddddcc165b colors back in DEBUG=2 [pr] (#9155) 2025-02-18 16:17:57 +08:00
George Hotz
6d62966bf7 add support for named rewrites [pr] (#9152) 2025-02-18 16:07:04 +08:00
George Hotz
caee42e8a6 Revert "name from uops [pr] (#9151)" (#9154) 
This reverts commit 28897be9a2.
 2025-02-18 16:06:44 +08:00
George Hotz
28897be9a2 name from uops [pr] (#9151) 2025-02-18 15:52:03 +08:00
George Hotz
a4dab3ec3f add name uop (#9149) 
* add name uop, TODO: refactor renderer to use

* renderer uses name uop

* fix tests

* render

* ptx
 2025-02-18 15:26:58 +08:00
George Hotz
2db8b4046a minor linearizer refactor to finalize in rewrite [pr] (#9148) 2025-02-18 12:42:22 +08:00