|
About the autograd category
|
|
0
|
3591
|
May 13, 2017
|
|
How to concatenate feature vectors from different extractors inplace
|
|
0
|
6
|
April 27, 2024
|
|
My network's weights get updated despite using torch.no_grad()
|
|
2
|
22
|
April 27, 2024
|
|
Is it safe to modify output's grad and return as input's grad?
|
|
4
|
36
|
April 27, 2024
|
|
How to make my faster GPU execute two batches while my other GPU does a batch?
|
|
0
|
21
|
April 26, 2024
|
|
`zero_grad` before `step` causes gradient explosion?
|
|
3
|
45
|
April 26, 2024
|
|
Meta learning weight update
|
|
0
|
22
|
April 26, 2024
|
|
Propagate gradient to parameters not directly involved in computation
|
|
9
|
169
|
April 25, 2024
|
|
RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.FloatTensor [200, 2]], which is output 0 of AsStridedBackward0, is at version 1401; expected version 1400 instead
|
|
3
|
117
|
April 24, 2024
|
|
Is it safe to set flag for single autograd.Function.backward call
|
|
3
|
55
|
April 24, 2024
|
|
Diverging gradients obtained through: 1) tensor.register_hook in a class vs 2) module.register_full_backward_hook outside of a class
|
|
0
|
99
|
April 24, 2024
|
|
Fixing the Training Function for an LSTM model built from scratch
|
|
4
|
48
|
April 24, 2024
|
|
What is the detailed operations of `torchfunc.jvp`
|
|
2
|
52
|
April 24, 2024
|
|
Getting gradients with respect to input features for a batched input
|
|
3
|
43
|
April 23, 2024
|
|
CUDA Out of Memory Error with jvp() in Large Model Training - Need Insights or Fixes
|
|
0
|
37
|
April 23, 2024
|
|
Mixing pytorch autograd with custom vjp for optimization
|
|
2
|
25
|
April 23, 2024
|
|
Loss.backward() call not updating loss
|
|
2
|
33
|
April 23, 2024
|
|
Out.backward() and weight access
|
|
1
|
34
|
April 23, 2024
|
|
Backpropagating w.r.t detached tensor
|
|
2
|
31
|
April 23, 2024
|
|
A question about autograd.Function
|
|
2
|
46
|
April 23, 2024
|
|
Backpropagate several error values
|
|
5
|
66
|
April 22, 2024
|
|
Finding the Diagonal of Hessian wrt Input for Vector-Valued Functions
|
|
1
|
49
|
April 21, 2024
|
|
Grad becomes nan for all parameters
|
|
7
|
103
|
April 20, 2024
|
|
How do I get `torch.onnx.export` to work with `aot_module`?
|
|
1
|
50
|
April 19, 2024
|
|
Differences between 'params.grad' and 'params.grad.data'
|
|
1
|
42
|
April 19, 2024
|
|
Model forward slows down when disabling gradients
|
|
5
|
85
|
April 19, 2024
|
|
Signed distance transform computation of predicted segmentation
|
|
6
|
4208
|
April 19, 2024
|
|
Intuition Behind Nesterov Update Rule for SGD w/ Momentum
|
|
0
|
41
|
April 19, 2024
|
|
Omit loss.backward for forward-only algos?
|
|
2
|
58
|
April 18, 2024
|
|
Set a portion of a tensor to have requires_grad=False
|
|
9
|
1419
|
April 18, 2024
|