|
Symmetric parametrization
|
|
2
|
37
|
June 1, 2025
|
|
Vmap runtime error
|
|
0
|
24
|
May 29, 2025
|
|
Autograd: Add VJP and JVP rules for aten::aminmax #151186
|
|
0
|
31
|
May 25, 2025
|
|
How to test new native function
|
|
1
|
31
|
May 23, 2025
|
|
Blown up gradients and loss
|
|
3
|
943
|
May 22, 2025
|
|
"RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.FloatTensor [64, 1]], which is output 0 of AsStridedBackward0, is at version 3; expected version 2 instead. Hint: the backtrace further a
|
|
9
|
31988
|
May 19, 2025
|
|
Removing Return Statement in Module Forward Causes 30+ms Backward Slowdown - Why?
|
|
1
|
22
|
May 15, 2025
|
|
Linear solver for sparse matrices
|
|
7
|
1420
|
May 15, 2025
|
|
Why does merging all loss in a batch make sense?
|
|
7
|
2460
|
May 14, 2025
|
|
Autograd and Temporary Variables
|
|
4
|
128
|
May 12, 2025
|
|
Batthacaryya loss
|
|
12
|
2939
|
May 12, 2025
|
|
Autograd FLOP Calculation with Higher Order Derivatives
|
|
3
|
177
|
May 9, 2025
|
|
Gradient of a mixed network's output with respect to ONE tensor
|
|
2
|
47
|
May 7, 2025
|
|
Segfault in autograd after using torch lightning
|
|
1
|
64
|
May 4, 2025
|
|
How to interactively debug pytorch backprop errors?
|
|
2
|
88
|
May 2, 2025
|
|
Gradient of a weight matrix returns NAN with respect to learnable parameter
|
|
2
|
43
|
May 1, 2025
|
|
Get loss, gradient and hessian in one go
|
|
1
|
95
|
April 30, 2025
|
|
How to clip the values of an optimizer?
|
|
3
|
51
|
April 30, 2025
|
|
Monitor optimizer step - Adam
|
|
1
|
76
|
April 30, 2025
|
|
Gradient of Tensor is Zero
|
|
4
|
117
|
April 28, 2025
|
|
Multiple forwards and comp graph building
|
|
2
|
58
|
April 23, 2025
|
|
My Discriminator model collapsed and always returns 1s
|
|
0
|
30
|
April 23, 2025
|
|
Autograd independently on entries of a single tensor
|
|
2
|
46
|
April 21, 2025
|
|
Optimize objective involving jacobian
|
|
1
|
72
|
April 21, 2025
|
|
Most efficient way to re-use grad computations in a layer which is a linear combination of linear layers
|
|
2
|
42
|
April 20, 2025
|
|
Simple use case: Compete per sample gradient with autograd
|
|
17
|
205
|
April 16, 2025
|
|
Is gradient flow lost when using Numpy?
|
|
3
|
139
|
April 15, 2025
|
|
Gradient computation performance
|
|
2
|
81
|
April 15, 2025
|
|
Autograd process isn't available when I profile via torch.profiler
|
|
3
|
32
|
April 15, 2025
|
|
Making autograd saved tensors hooks specific to certain arguments
|
|
7
|
113
|
April 14, 2025
|