timbmg/Sentence-VAE
PyTorch Re-Implementation of "Generating Sentences from a Continuous Space" by Bowman et al 2015 https://arxiv.org/abs/1511.06349
Python
Issues
- 3
Confusion about the batch in model.py(I think batch should be the second dimension.)
#15 opened by WenXiuxiu - 0
bi-directional gru
#35 opened by Binyun-Z - 1
imputation
#32 opened by Piao-Hu - 1
Wrong usage of Bidirectional GRU
#30 opened by sharan21 - 0
Pre-trained model?
#29 opened by flarn2006 - 1
- 2
Maybe a bug in constructing hidden states.
#20 opened by leix28 - 0
inference
#28 opened by safakkbilici - 0
Change in ELBO during training
#27 opened by soumyasanyal - 2
maybe a bug in the train.py
#18 opened by linzhk - 2
Index Error during Inference
#24 opened by tnhaider - 4
Inference edge-case bug.
#14 opened by MBAnslow - 0
Optimiser SGD leads to KL vanishing
#25 opened by ZhuohanX - 2
Unable to download the dataset
#23 opened by nanguoshun - 1
maybe a bug in the kl_anneal_function
#19 opened by linzhk - 0
Discrete latent variable
#16 opened by izaskr - 2
Repeated content
#7 opened by nguyenvo09 - 4
Question about the inference function
#13 opened by vikigenius - 1
How to get the init hidden state of LSTM?
#10 opened by JianLiu91 - 1
Word Dropout not working
#11 opened by dhanajitb - 0
question on inputs to encoder and decoder
#9 opened by zshihang - 1
Missing requirements.txt
#8 opened by gagb - 4
Zero-dimensional tensor
#4 opened by gsramsu - 4
word dropout or word embedding dropout?
#5 opened by mengxuehu - 2
The input of decoder seems not right
#3 opened by psubnwell - 5
which version of Pytorch, why i take the project to run,there are many problem? such as this code"hidden = hidden[:,running_seqs]" wish you can give me some advise, thanks
#1 opened by KJ-Lin - 1
Error due to invalid data types
#2 opened by Deepayan137