TorchInductor: a PyTorch-native Compiler with Define-by-Run IR and Symbolic Shapes
|
|
46
|
58483
|
July 29, 2024
|
How to set wrap function using TorchDynamo graph capture?
|
|
4
|
321
|
June 28, 2024
|
Understanding torch.fx.traceback.preserve_node_meta()
|
|
0
|
88
|
July 26, 2024
|
Supporting Dynamo in Python 3.12
|
|
0
|
525
|
July 26, 2024
|
Inlining a custom triton kernel
|
|
0
|
115
|
July 22, 2024
|
Compiled autograd with custom ops error
|
|
1
|
204
|
July 19, 2024
|
AOTAutograd incorrect lowering composite ops in inference_mode
|
|
3
|
231
|
July 17, 2024
|
[RFC] Performance profiling at scale with detailed NVTX annotations
|
|
0
|
252
|
July 10, 2024
|
State of symbolic shapes branch
|
|
96
|
30719
|
July 7, 2024
|
Why PyTorch does not need a new standardized operator set
|
|
3
|
848
|
July 3, 2024
|
Understanding the lifecycle of compiled code cache entry
|
|
2
|
236
|
July 2, 2024
|
Semantics of None in aten.index
|
|
1
|
160
|
July 2, 2024
|
Fixing torch.compile Reference Leaks (Automatic Deletion of Dynamo Code Objects)
|
|
0
|
503
|
July 2, 2024
|
Reverse Fusion of Node Pairs in Scheduler
|
|
0
|
147
|
June 14, 2024
|
Custom C++ External Kernel for TorchInductor
|
|
2
|
226
|
June 10, 2024
|
PyTorch/XLA 2.3 dev update
|
|
0
|
691
|
May 29, 2024
|
User-defined Kernels vs. `torch.library` custom op
|
|
1
|
287
|
May 22, 2024
|
Different points at which fusion occurs?
|
|
8
|
343
|
May 21, 2024
|
PyTorch Sparse(GNN) Compiler RFC
|
|
28
|
2034
|
May 21, 2024
|
How to replace a callable with an opaque python callable in an Inductor pass?
|
|
6
|
246
|
May 15, 2024
|
How To Bring Compile Time Down to Zero: Our Plans and Direction (May 14th Edition)
|
|
0
|
1115
|
May 15, 2024
|
When does the inductor code run?
|
|
5
|
427
|
May 15, 2024
|
What is the mental model for torch.compile to deal with stateful objects in general?
|
|
3
|
267
|
May 13, 2024
|
How can I dump the prims IR, triton code, and ptx code when using torch.compile()
|
|
2
|
826
|
May 6, 2024
|
Torch.compile support for Python 3.12 completed
|
|
0
|
1953
|
May 3, 2024
|
Performance Comparison between Torch.Compile and APEX optimizers
|
|
1
|
1615
|
May 1, 2024
|
AOT module decompositions
|
|
0
|
255
|
April 25, 2024
|
Understanding CUDAGraph Trees
|
|
4
|
875
|
April 11, 2024
|
State of PT2 OSS Issues: Q1 2024
|
|
0
|
245
|
April 10, 2024
|
Inductor CUDA Backend
|
|
1
|
624
|
April 4, 2024
|