OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 00:21:07 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
f7365eca90
pytorch/torch/_inductor
History
XiaobingSuper 4ca2fc485c inductor(CPU): add Conv+binary+unary fusion filter (#90259) 
For Conv+binary+unary fusion, we only support conv+add+relu, this PR adds a such check to fix TIMM failed models.
TODO: enable more Conv+binary+unary fusion to improve TIMM models' performance.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/90259
Approved by: https://github.com/EikanWang, https://github.com/jgong5, https://github.com/jansel
2022-12-12 06:04:55 +00:00
..
codegen Emit torch.cuda.synchronize() after every kernel call in inductor (#90472) 2022-12-12 04:35:10 +00:00
triton_ops [Dynamo] Fix llvm target for meta schedule & add torch to tvm ndarray helper func (#90214) 2022-12-07 19:23:56 +00:00
__init__.py
codecache.py Revert "Dynamo, FX, Inductor Progress Bars (#88384)" 2022-12-09 16:32:25 +00:00
compile_fx.py Always compile tiny graphs with AOTAutograd (#89775) 2022-12-08 03:41:29 +00:00
config.py Emit torch.cuda.synchronize() after every kernel call in inductor (#90472) 2022-12-12 04:35:10 +00:00
cuda_properties.py Fix TorchInductor benchmarking in fbcode (#88689) 2022-11-09 18:13:06 +00:00
debug.py [pt2] Reset dynamo log level when exiting inductor debug context (#90473) 2022-12-12 04:39:37 +00:00
decomposition.py [Inductor] GEMM Shape Padding Optimization (#90425) 2022-12-09 22:48:02 +00:00
dependencies.py Added utility to count memory reads/written in Inductor (#89203) 2022-11-19 04:18:26 +00:00
exc.py
graph.py Revert "Dynamo, FX, Inductor Progress Bars (#88384)" 2022-12-09 16:32:25 +00:00
ir.py Revert "Dynamo, FX, Inductor Progress Bars (#88384)" 2022-12-09 16:32:25 +00:00
lowering.py [inductor][Reland] Use decomposition for _to_copy (#90494) 2022-12-09 16:51:50 +00:00
metrics.py [Inductor] Add test for Scheduler fusions (#90014) 2022-12-07 01:33:25 +00:00
overrides.py inductor(CPU): add Conv+binary+unary fusion filter (#90259) 2022-12-12 06:04:55 +00:00
scheduler.py Emit torch.cuda.synchronize() after every kernel call in inductor (#90472) 2022-12-12 04:35:10 +00:00
sizevars.py Keep track of source name on all allocated SymInts (#90295) 2022-12-10 13:17:34 +00:00
utils.py pad low precision matmuls when requested (#90235) 2022-12-06 04:13:24 +00:00
virtualized.py Revert "Dynamo, FX, Inductor Progress Bars (#88384)" 2022-12-09 16:32:25 +00:00