autograd


Topic Replies Activity
Model weight cannot update when gradients are computed from distributed machines 4 November 13, 2019
Add two optimizers in single model for two different path 5 November 13, 2019
Parameter vs tensor.requires_grad = True 3 November 12, 2019
Does share_memory model is lock free? 2 November 12, 2019
Some confusion about the grad of torch.sign() 6 November 12, 2019
Which implementation performs mini-batch gradient descent correctly? 1 November 12, 2019
When to set up autograd in nn module 3 November 11, 2019
Running a function on GPU 2 November 11, 2019
Unction CudnnConvolutionBackward returned an invalid gradient at index 1 - expected type torch.FloatTensor but got torch.cuda.FloatTensor 10 November 11, 2019
[Memory problem] Replace input by another tensor in the forward pass 10 November 8, 2019
Runtime Error: Function MulBackward0 returned an invalid gradient at index 1 - expected type torch.FloatTensor but got torch.cuda.FloatTensor 3 November 11, 2019
Gradients are always zero 4 November 10, 2019
How to backpropagate a black box generated cost value 6 November 9, 2019
Using Variable to update nn.Parameter and keep gradient 1 November 9, 2019
GRU included CNN for text Generator 1 November 9, 2019
Code supposed to be similar but different result 6 November 8, 2019
Gradient with respect to parameters that update model parameters? 3 November 7, 2019
How to get input&output tensor of a module in backward process 6 November 7, 2019
Understanding autograd calculation of backprop 4 November 7, 2019
Backward() None issue with autograd 4 November 7, 2019
Does calling backward on two different variables affect how gradients are calculated? 2 November 7, 2019
Ctx.needs_input_grad behaviour 14 November 7, 2019
Is there a way to parallelize meta-update for MAML? 1 November 7, 2019
Memory blows up when evaluating model even with ‘with torch.no_grad' and 'model.eval' 3 November 7, 2019
Norm of a Tensor 5 November 6, 2019
PyTorch gradient differs from JAX 2 November 5, 2019
Inplace operations for custom layers complicating backpropagation 6 November 5, 2019
Status of register_backward_hook 8 November 5, 2019
Why there exists no grad after tensor is reshaped? 2 November 5, 2019
Backward error : 'NoneType' object has no attribute 'data' 4 November 4, 2019