Topic Replies Activity
Building libtorch for IBM PLX
C++
4 August 20, 2019
Backwards hook dimensions unexpected 1 August 20, 2019
Anyway to parallelize this function? 2 August 20, 2019
How to access input/output activations of a layer given its parameters names? 3 August 20, 2019
Dataloader freezes when num_workers>0 on windows without GPU 1 August 20, 2019
Using DataParallel with functional API 1 August 20, 2019
Difference in Model Performance when using Validation set/ Testing set 1 August 20, 2019
Cutoffs for Adaptive Softmax 1 August 20, 2019
How to reset model weights to effectively implement crossvalidation? 1 August 20, 2019
How to use torch.topk() to set non-topk values of a tensor to zero? 4 August 20, 2019
Model cannot learn useful features - test loss is always much lower than train 1 August 8, 2019
AttributeError: 'Sequential' object has no attribute 'weight' 2 August 20, 2019
Custom LSTM returns nan
jit
3 August 20, 2019
Is there a parser for strings of Sequential Models? 2 August 20, 2019
Broken API in libtorch 1.2
C++
3 August 20, 2019
Model.eval() Error -- 'bool' object is not callable 3 August 20, 2019
Torch.nn model cannot pad a 1D tensor 1 August 20, 2019
RuntimeError: shape '[-1, 400]' is invalid for input of size 6 August 20, 2019
Integrating PackedSequence with a custom recurrent model 1 August 20, 2019
How to shuffle data in Dataloader
C++
4 August 20, 2019
Error for affine_grid_generator when using 5D tensor 2 August 20, 2019
nn.Transformer explaination
nlp
4 August 20, 2019
Full Log Poisson Loss for PyTorch 2 August 20, 2019
Backpropagate through input data 4 August 20, 2019
[Please Help] in-place operation in module 2 August 20, 2019
Blas version the pytorch uses 2 August 20, 2019
Current way to compile from source 4 August 20, 2019
Passing input to the Linear model 2 August 20, 2019
Adding dropout after training 4 August 20, 2019
Indexing and updating with tensors 2 August 20, 2019