Freezing layers when using torch.compile
|
|
2
|
174
|
February 18, 2024
|
How can I get the exact torch operator of each node?
|
|
0
|
152
|
February 15, 2024
|
TypeError when exporting model with dynamo_export
|
|
0
|
155
|
February 15, 2024
|
How to get the computation graph and pass it to C++?
|
|
3
|
1639
|
February 15, 2024
|
Decomposition error with `scaled dot product flash attention`
|
|
0
|
133
|
February 12, 2024
|
How can I see the source of guards?
|
|
0
|
153
|
February 9, 2024
|
Compile source code, code navigation
|
|
2
|
189
|
February 9, 2024
|
Developer documentation for PyTorch 2.x compiler
|
|
12
|
552
|
February 6, 2024
|
Using torch.compile twice on a model on the same machine, is there a cache of optimized operations?
|
|
0
|
146
|
February 3, 2024
|
BackendCompilerFailed: backend='inductor' raised: RuntimeError: Triton Error [CUDA]: device kernel image is invalid
|
|
2
|
570
|
February 2, 2024
|
Does the order of graph_module.graph.nodes affect performance?
|
|
0
|
156
|
February 2, 2024
|
Torch.cuda.is_available() returns false even after accessing cuda toolkit 11.8
|
|
2
|
267
|
February 1, 2024
|
Why my torch.compiled model run so slow。
|
|
0
|
314
|
January 31, 2024
|
Torch compile on cpu slow
|
|
3
|
261
|
January 31, 2024
|
`compiled_autograd`
|
|
0
|
161
|
January 26, 2024
|
How to control or tell torch.compile for training or inference?
|
|
0
|
193
|
January 28, 2024
|
Choice of torch.compile vs. triton
|
|
6
|
1011
|
January 24, 2024
|
Output runnable backwards _inductor graph `torch.compile`
|
|
0
|
153
|
January 23, 2024
|
Replacing math operator with custom layer inside existing model
|
|
0
|
174
|
January 23, 2024
|
The CUDA Graph is empty
|
|
2
|
222
|
January 23, 2024
|
Prologue fusion
|
|
0
|
226
|
January 23, 2024
|
bin/python: relocation error: /lib/python3.11/site-packages/torch/lib/libcudnn_cnn_infer.so.8: symbol _ZN15TracebackLoggerC1EPKc version libcudnn_ops_infer.so.8 not defined in file libcudnn_ops_infer.so.8 with link time reference
|
|
1
|
263
|
January 23, 2024
|
How to only make a change to a submodule and then update Pytorch
|
|
3
|
263
|
January 22, 2024
|
C++/Cuda / aot.compile and cuda graph
|
|
0
|
163
|
January 22, 2024
|
Cannot Find ptxas
|
|
6
|
552
|
January 21, 2024
|
How to customize leaf node in dynamo trace?
|
|
0
|
222
|
January 19, 2024
|
How can Pytorch 2.x support different backends?
|
|
2
|
185
|
January 19, 2024
|
Graph breaks or No graph breaks in torchdynamo
|
|
3
|
1401
|
January 18, 2024
|
Where to PrimTorch's operator list?
|
|
2
|
201
|
January 18, 2024
|
Ignore a inner function with @torch.compile
|
|
1
|
294
|
January 18, 2024
|