practical-pytorch

practical-pytorch copied to clipboard

The attention mechanism is not the original attention mechanism in the paper

7

In the tutorial, I find that the "attention" mechanism is a fake attention, since the calculated attention weights have no relationship with the encoder output vectors. The implementation in the...

rk2900

Train and Test Split

1

In the Char RNN classification I had some questions: 1. Why have not split the dataset into train and test/validation set to check the performance ? 2. How can we...

rjtmehta99

Convert 0 dimension tensor to number.

pritoms

chuanky

In the train.py script, Line 45: "return output, loss.data[0]", ---> "return output, loss.item()" is suggested.

jinghuayao

bjyx-star

ziyuanca

About seq2seq-translation-batched.py RuntimeError

1

there was error occured: ``` in score(self, hidden, encoder_output) 35 elif self.method == 'general': 36 energy = self.attn(encoder_output) ---> 37 energy = hidden.dot(energy) 38 return energy 39 RuntimeError: 1D tensors...

xiangningOvo

Can seq2seq model support the new sentence tranlslation?

5

Why does the KeyError occurs?I can not understand. Can you help me?Thank you  

quoniammm

Fix a bug in evaluate()

Tad-Done