Generating 2D Tensor by Indexing 3D Tensor with 2D LongTensor (1)
Getting rid of maxpooling layer causes running cuda out memory error pytorch (2)
Is this a bug in torch.nn.EmbeddingBag? (9)
Is there a masking method for CNN input of variable size within a batch? (5)
Where is torch.tensor() actually defined? (2)
RNN Loss Calculation Quitting (4)
Custom data loading process become too slow (2)
How to load sparse matrix as training input (1)
C++ API to choose GPU from multiple gpus (1)
Balanced trainLoader (3)
Cross entropy for soft label (2)
Weight initilzation ( 2 3 ) (59)
How to assign a slice of a tensor to another tensor in-place in Caffe2? (2)
Regarding updating the weight of a model (7)
Convolution that takes a function as kernel (7)
Discrete Cosine Transform Using `torch.rfft` (7)
torch.LongTensor doesn't initialize to 0 above certain size (3)
How to explain huge GPU RAM usage? (5)
Does Model learns even when Validation error seems to be saturated (1)
[solved] Discrete cosine transform implementation in pytorch? (2)
How a C++ program call a network written in PyTorch (1)
Weight decay together with L2 regularization baked into loss function (3)
Repeat a nn.Parameter for efficient computation (9)
Can't compile PyTorch 0.4.1 with CUDA support on Linux; commands not found (3)
When to use repeat instead of expanse? (2)
Question about group convolution (4)
How to load part of model when the model contains more layers than the save weights? (2)
Should custom Dataset always return tensors? (9)
ScatterAssign is CPU only? (1)
Using register_forward_hook for recurrent networks (1)