Shapes of FakeTensors after graph transformations on ATen IR
|
|
6
|
58
|
March 19, 2025
|
Torch.compile(mode="max-autotune") produces different inference result from eager mode — is this expected?
|
|
6
|
75
|
March 19, 2025
|
Triton CodeGen for torch linear operator
|
|
0
|
7
|
March 19, 2025
|
How to solve the graph break happen in torch.compile
|
|
11
|
398
|
March 18, 2025
|
The selectable configs for irregular shape when using triton.autotune
|
|
0
|
17
|
March 18, 2025
|
Torch.compile CudaGraph creation & downstream systems
|
|
2
|
35
|
March 18, 2025
|
Debugging slow TorchDynamo Cache Lookup
|
|
1
|
33
|
March 18, 2025
|
How to ensure that auto-generated Triton Kernel is executed
|
|
0
|
14
|
March 13, 2025
|
How to let torch.compile only target GPU
|
|
0
|
43
|
March 13, 2025
|
Flex attention raises unnecessary assertion error
|
|
1
|
60
|
March 12, 2025
|
`cuda.h' missing during torch.compile in new environment
|
|
3
|
252
|
March 7, 2025
|
Why is inductor re-recording my function with every call
|
|
3
|
47
|
March 6, 2025
|
PyTorch v.2.3.1 unable to configure
|
|
1
|
223
|
March 4, 2025
|
Torch._check failing with torch compile
|
|
1
|
76
|
March 1, 2025
|
How to make TorchDynamoc decompose torch.nn.functional.scaled_dot_product_attention?
|
|
1
|
45
|
February 27, 2025
|
Mapping between TorchDynamo compiled functions and FX GraphModules
|
|
2
|
42
|
February 27, 2025
|
PyTorch 2.5.1 -fstack-protector-strong does not result in stack-protected libpytorch.so
|
|
1
|
44
|
February 25, 2025
|
Ways to pass compilation target device to dynamo backend
|
|
2
|
390
|
February 20, 2025
|
Converting model that has dynamic control flow to load in C++
|
|
0
|
21
|
February 19, 2025
|
Seeking insights: Strange torch.compile() behavior with RegionViT in inference_mode()
|
|
0
|
30
|
February 15, 2025
|
Too many compile workers and huge RAM consumption
|
|
3
|
59
|
February 14, 2025
|
Dynamo.explain() output
|
|
0
|
50
|
February 13, 2025
|
How does torch.fx work with activation checkpointing?
|
|
2
|
773
|
February 13, 2025
|
Getting Triton to generate all kernels
|
|
13
|
2714
|
February 11, 2025
|
Generate Triton kernels for CPU
|
|
5
|
959
|
February 11, 2025
|
Build PyTorch without AVX/AVX2 or CUDA for old systems
|
|
0
|
125
|
February 11, 2025
|
Torch IR support for torch.exported graph
|
|
0
|
39
|
February 10, 2025
|
Does dynamo trigger real kernel execution?
|
|
3
|
139
|
February 8, 2025
|
Support for torch.equal when exporting with Dynamo
|
|
3
|
38
|
February 7, 2025
|
[torch.export] How to disable sym_size_int?
|
|
2
|
125
|
February 7, 2025
|