|
About the autograd category
|
|
0
|
1715
|
May 13, 2017
|
|
What is the difference between torch.vmap and functorch?
|
|
1
|
22
|
September 23, 2021
|
|
RuntimeError: Function MmBackward returned an invalid gradient at index 0 - got [2048, 256] but expected shape compatible with [2048, 265]
|
|
0
|
10
|
September 23, 2021
|
|
What is the storage structure of computation graph, what is freed after backward() and how it is freed?
|
|
2
|
24
|
September 23, 2021
|
|
Backprop through generator to update latent
|
|
0
|
14
|
September 23, 2021
|
|
Gradient Checkpointing not backpropping in GNN
|
|
1
|
105
|
September 23, 2021
|
|
Trying to find an angle (use of trigonometric functions); optimizer not updating
|
|
3
|
1868
|
September 22, 2021
|
|
Accumulate gradient with nonlinear global function with small memory footprint
|
|
1
|
32
|
September 22, 2021
|
|
Effect of tensor modification in forward pass on gradient calculation
|
|
3
|
47
|
September 22, 2021
|
|
Custom loss function leading to loss of gradients
|
|
1
|
21
|
September 21, 2021
|
|
Why next_functions[0][0] is called with no loss on a branch
|
|
1
|
29
|
September 21, 2021
|
|
Force gradients computation even in inference mode
|
|
3
|
34
|
September 21, 2021
|
|
Pytorch calculation precision
|
|
1
|
27
|
September 20, 2021
|
|
Loss function that uses the output to calculate index which is used to get value from an array and then loss is calculated
|
|
1
|
25
|
September 20, 2021
|
|
Gradient w.r.t target values
|
|
5
|
32
|
September 19, 2021
|
|
Is autograd storing the activations somewhere?
|
|
4
|
65
|
September 18, 2021
|
|
Segmentation fault when using torch.profiler
|
|
0
|
19
|
September 18, 2021
|
|
Slow batch_gradient
|
|
3
|
33
|
September 17, 2021
|
|
How can you do memory efficient MAML?
|
|
6
|
72
|
September 17, 2021
|
|
RuntimeError: expected scalar type Long but found Float with Custom Data
|
|
3
|
32
|
September 17, 2021
|
|
How to convert model's parameters to tensor and keep grad?
|
|
4
|
92
|
September 15, 2021
|
|
Computational graph's working during updating parameters
|
|
0
|
27
|
September 15, 2021
|
|
How to prevent leak when using .cuda() on temporary view and replacing original variable?
|
|
2
|
28
|
September 15, 2021
|
|
Computing gradient inside torch.autograd.Function
|
|
2
|
29
|
September 14, 2021
|
|
[Adding functionality] Hessian and Fisher Information vector products
|
|
13
|
2988
|
September 14, 2021
|
|
What happens when we use torch.no_grad() in the middle of forward pass?
|
|
3
|
39
|
September 14, 2021
|
|
No inf checks were recorded for this optimizer error for custom loss
|
|
1
|
23
|
September 14, 2021
|
|
Autograd yields nan even if nan input is independant of output
|
|
3
|
44
|
September 13, 2021
|
|
Backprop machinery
|
|
2
|
55
|
September 13, 2021
|
|
One of the variables needed for gradient computation has been modified by an inplace operation for this specific piece of code
|
|
2
|
27
|
September 13, 2021
|
