practical-pytorch
practical-pytorch copied to clipboard
spro
Go to https://github.com/pytorch/tutorials - this repo is deprecated and no longer maintained
Not sure if only on my platform (win10 python 3.6), but couldn't run the char rnn classification without this fix
As beam search is widely used in machine translation, could you support beam search sampling and evaluate the model?
Im'new guy in dl and pytorch. i read your tutorial https://pytorch.org/tutorials/intermediate/char_rnn_generation_tutorial.html but i wonder ,why you inherit nn.module as a Rnn,but you dont you nn.lstm etc. why? Is nn.lstm is...
In the RNN code here: https://pytorch.org/tutorials/intermediate/char_rnn_classification_tutorial.html, ``` self.i2h = nn.Linear(input_size + hidden_size, hidden_size) self.i2o = nn.Linear(input_size + hidden_size, output_size) ``` there seem to be no activation present, tanh or Relu...
I am wondering if the attention score calculation could be vectorized to run per batch instead of running two loops (batch size x encoder time steps) The sections of code...
``` Training for 2000 epochs... Traceback (most recent call last): File ".\train.py", line 64, in loss = train(*random_training_set(args.chunk_len)) File ".\train.py", line 49, in train loss += criterion(output, target[c]) File "C:\Users\leond\AppData\Local\Continuum\Anaconda3\lib\site-packages\torch\nn\modules\module.py",...
Hi, I am implementing seq2seq attention mechanism with using different input and output (not machine translation). But when I apply your code into mine, there's dimension error.. so I want...
In this https://github.com/spro/practical-pytorch/blob/master/seq2seq-translation/seq2seq-translation.ipynb, why do we need to feed the last_context as an rnn input? ```python # Combine embedded input word and last context, run through RNN rnn_input = torch.cat((word_embedded,...
shakespeare.txt file which is used in char-run-generation is not present in the data folder.
