|
About the autograd category
|
|
0
|
3985
|
May 13, 2017
|
|
Forward autodiff : Multiplying by python float changes the dual dtype in some situations
|
|
4
|
60
|
August 13, 2025
|
|
Skip the calculation of gradients for some layer after forward pass
|
|
1
|
25
|
August 9, 2025
|
|
PyTorch `torch.no_grad` vs `torch.inference_mode`
|
|
2
|
26984
|
August 5, 2025
|
|
RuntimeError when exponentiating matrix using eigendecomposition
|
|
1
|
22
|
August 5, 2025
|
|
Higher library and create/reatain graph error
|
|
1
|
36
|
August 5, 2025
|
|
When writing a torch.autograd.Function, is there some way for the .backward() to know with respect to which input are the gradients being requested?
|
|
3
|
64
|
July 31, 2025
|
|
Nested forward mode AD not supported (jacobian)— what operations can trigger this?
|
|
1
|
24
|
July 28, 2025
|
|
Verifying grad_x after autograd double backprop
|
|
1
|
24
|
July 24, 2025
|
|
Custom Physics-Informed Loss Function Not Converging for PINN Thermodynamics Model
|
|
0
|
32
|
July 22, 2025
|
|
Can't vmap autograd.grad over outputs
|
|
10
|
140
|
July 16, 2025
|
|
Autograd function returns tensor views
|
|
1
|
40
|
July 16, 2025
|
|
Backward through the graph a second time
|
|
2
|
45
|
July 12, 2025
|
|
LSTM model weight gradients are all zero
|
|
3
|
41
|
July 11, 2025
|
|
Custom loss and its gradient
|
|
1
|
40
|
July 10, 2025
|
|
Why does autograd.backward go one edge further than `inputs`?
|
|
3
|
45
|
July 8, 2025
|
|
Implementing calculation of the Laplacian
|
|
6
|
5265
|
July 7, 2025
|
|
Efficient Computation of Second-Order Gradients
|
|
1
|
30
|
July 7, 2025
|
|
Tensor hooks for gradients from each path in the autograd graph
|
|
4
|
70
|
July 5, 2025
|
|
Memory bloat in gradient descent model
|
|
4
|
67
|
July 2, 2025
|
|
In-place modification in backward not allowed with torch.compile
|
|
1
|
22
|
July 1, 2025
|
|
Multi-discriminator GAN: No inf checks were recorded for this optimizer
|
|
2
|
33
|
June 25, 2025
|
|
Variable versioning error at backward
|
|
1
|
30
|
June 21, 2025
|
|
How to know if gradients backpropagate through a pytorch function?
|
|
1
|
550
|
June 20, 2025
|
|
How pytorch treat with inplace operation in backward
|
|
1
|
47
|
June 17, 2025
|
|
[SDPA] RTX5080 is different from CPU calculation result in backward with long seq
|
|
0
|
26
|
June 17, 2025
|
|
[BUG] RTX5080： Function 'MmBackward0' returned nan values in its 0th output.
|
|
2
|
43
|
June 16, 2025
|
|
Broken autograd momentum link
|
|
1
|
34
|
June 16, 2025
|
|
JVP and checkpointing
|
|
1
|
52
|
June 16, 2025
|
|
Constant Predictions in Non-Linear Model Despite Training Progress
|
|
2
|
47
|
June 15, 2025
|