|
How to re-decompose operator in Inductor pass
|
|
0
|
335
|
December 6, 2023
|
|
How to trace torch.autograd.backward or torch.autograd.grad?
|
|
3
|
1296
|
December 5, 2023
|
|
Arithmatic expression of prims operator
|
|
0
|
239
|
December 3, 2023
|
|
Call for backward compatability to enable users to understand and adapt to pytorch compiler
|
|
2
|
722
|
November 18, 2023
|
|
Torch.compile() tech talks at PTC'23
|
|
0
|
653
|
October 31, 2023
|
|
Torch.compile with AOT Autograd can be debugged now!
|
|
1
|
882
|
October 31, 2023
|
|
Torch.compile can be debugged now!
|
|
2
|
2273
|
October 24, 2023
|
|
Interactively explore what torch.compile does to your code!
|
|
0
|
880
|
October 15, 2023
|
|
PyTorch/XLA 2.1 Release Dev Update
|
|
1
|
1756
|
October 13, 2023
|
|
Is Dynamo enabled per-thread or per-process?
|
|
2
|
1072
|
October 7, 2023
|
|
A minimal working example of standalone usage for dynamo eval_frame
|
|
2
|
877
|
September 29, 2023
|
|
TorchInductor Update 6: CPU backend performance update and new features in PyTorch 2.1
|
|
0
|
2163
|
September 22, 2023
|
|
Why Dynamo fails to capture the computation graph in this function?
|
|
1
|
607
|
September 14, 2023
|
|
A proposal to make Dynamo more understandable to users
|
|
3
|
1716
|
September 5, 2023
|
|
I build a decompiler to convert bytecode generated by dynamo into readable source code!
|
|
4
|
879
|
August 29, 2023
|
|
Supporting Dynamo in Python 3.11 - CPython Frame Evaluation
|
|
0
|
1521
|
August 8, 2023
|
|
How are guards installed on frames that are transient objects?
|
|
9
|
900
|
August 6, 2023
|
|
How to capture NCCL communication ops in FakeTensorMode?
|
|
3
|
977
|
August 3, 2023
|
|
Supporting Dynamo in Python 3.11 - NULL
|
|
0
|
1796
|
July 24, 2023
|
|
Can we add a default backend when OpenMP is not available?
|
|
5
|
954
|
July 21, 2023
|
|
Questions about Inductor code generation for gemm on Nvidia device
|
|
4
|
1743
|
July 21, 2023
|
|
[Fatal Bug] changed nn.Module.training does not trigger recompilation
|
|
0
|
439
|
July 20, 2023
|
|
What is the relation between PrimTorch and Decompostions?
|
|
6
|
2461
|
July 7, 2023
|
|
Why is silu_backward decomposed in fx graph trace?
|
|
4
|
1629
|
June 21, 2023
|
|
Dynamo allow in graph fail on tensor.view(dtype)
|
|
1
|
677
|
June 17, 2023
|
|
Backward module does not contains weight's gradients calculation
|
|
0
|
502
|
June 16, 2023
|
|
Lengths' argument should be a 1D CPU int64 tensor, but got 1D meta Long tensor
|
|
1
|
616
|
June 12, 2023
|
|
Inductor updates
|
|
0
|
1316
|
June 1, 2023
|
|
Add a new compiler backend for jit.trace
|
|
1
|
693
|
May 25, 2023
|
|
How to differentiate target devices in dynamo backends?
|
|
1
|
701
|
May 24, 2023
|