autograd


About the autograd category (1)
The new .to(device) doesn't work with nn.DataParallel (2)
Variable grad is always None when extending autograd (8)
Detach, no_grad and requires_grad (1)
Loss.backward() failed at second round for customized loss function (1)
Memory Leak in LSTM (6)
Repeat function probably freezing my computer (1)
How to get around the RuntimeError: in-place operations can be only used on variables that don't share storage with any other variables (7)
Torch.utils.checkpoint.checkpoint (1)
Delay DataParallel grad gathering (1)
Retain_graph error (2)
Multi headed losses (1)
Backward for negative log likelihood loss of MultivariateNormal (in distributions) (2)
_sparse_mask attribute in sparse gradients computed with custom torch.autograd.Function (1)
Beam Search: RuntimeError: element 0 of variables does not require grad (1)
RuntimeError: element 0 of variables does not require grad (9)
Sparse Embedding GPU memory issue (1)
How to compute Jacobian matrix in PyTorch? (7)
In-place operations can be only used on variables that don't share storage with any other variables, but detected that there are 2 objects sharing it (7)
Implementation of Autograd for Sparse Tensor (1)
Understanding load_state_dict() effect on computational graph (9)
Custom Top-eigenvector Function (4)
Autograd happens when changes the data of Variable with `a[:,:,1,1]=2`? (4)
What if you still pass frozen parameters to optimizer? (2)
Best way to convolve on different channels with a single kernel? (5)
Strange error when running my model in double data type (2)
Backpropagation step is too slow for large embeddings (1)
Now that pytorch can autograd, why we still need to define backward when defining customer function (19)
Issue with loss exploding after a random number of epochs (6)
How does loss know what to do with batches? (does it broadcast?) (2)