|
About the torch._dynamo category
|
|
0
|
153
|
December 6, 2022
|
|
ValueError: --optim adamw_torch_fused with --fp16 requires PyTorch>2.0
|
|
12
|
85
|
March 23, 2023
|
|
"PTX .version 7.4 does not support .target sm_89" Again with latest Pytorch2.0, 4090 with CUDA12 and Ubuntu Linux 22.04
|
|
9
|
160
|
March 22, 2023
|
|
`torch.compile` + `torch.no_grad` not working for Mask R-CNN
|
|
2
|
72
|
March 22, 2023
|
|
[TorchDynamo] Too many function args in a fx graph
|
|
2
|
91
|
March 22, 2023
|
|
Pytorch 2.0 compile warning meaning
|
|
1
|
119
|
March 22, 2023
|
|
How to get whole graph when subgraph_compile triggle
|
|
1
|
69
|
March 22, 2023
|
|
UserDefinedObjectVariable should not be break renson
|
|
1
|
40
|
March 22, 2023
|
|
Triton Error [CUDA]: an illegal memory access was encountered
|
|
9
|
182
|
March 22, 2023
|
|
PyTorch 2.0 torch.compile works on CPU but not GPU
|
|
2
|
120
|
March 17, 2023
|
|
Question about graph break related to RETURN_VALUE and INLINE function call
|
|
1
|
57
|
March 17, 2023
|
|
Aot_eager mode segfault during bert training
|
|
0
|
35
|
March 14, 2023
|
|
The benchmark test for dynmao inductor runs stuck
|
|
1
|
41
|
March 2, 2023
|
|
RuntimeError: Internal Triton PTX codegen error, PTX .version 7.4 does not support .target sm_89
|
|
11
|
444
|
February 21, 2023
|
|
Compromised PyTorch-nightly dependency chain between December 25th and December 30th, 2022
|
|
5
|
326
|
January 10, 2023
|
|
torch._dynamo.exc.Unsupported: Tensor.backward
|
|
4
|
178
|
December 30, 2022
|
|
Torch2.0 torch.compile(model) error:
|
|
11
|
1253
|
December 23, 2022
|
|
Matmul slow for small matrices
|
|
2
|
167
|
December 23, 2022
|
|
Using torch compile with autocast
|
|
1
|
523
|
December 16, 2022
|