ctallec world-models issues

ctallec / world-models

Reimplementation of World-Models (Ha and Schmidhuber 2018) in pytorch

MIT License

568 stars 131 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Worker dying issue with controller training

#43 rohanb2018 opened 2 years ago
0
fix logsigma naming

#42 seolhokim opened 2 years ago
0
MDRNN doesn't train properly on carracing?

#41 hanbrianlee opened 3 years ago
0
problem about training VAE

#40 Qingcf closed 3 years ago
0
MDRNN losses extremely low due to numerical instability?

#39 parthjaggi opened 3 years ago
0
adding reference

#38 cjto2000 closed 3 years ago
1
the train_controller always break off when trainning about 15min

#37 longfeizhang617 opened 3 years ago
0
Controller Input

#36 zhefan opened 3 years ago
1
issue about gmm_loss

#35 UPUPGOO closed 4 years ago
2
Multiprocessing very slow

#34 TrentBrick opened 4 years ago
2
The definition of GMM linear layer may wrong? Or I have missed something?

#33 Reinelieben opened 4 years ago
4
Training the controller and getting stuck in local minima

#32 SynecFD closed 4 years ago
5
a multi-process problem in the controller

#31 sfrazavi opened 4 years ago
0
Fixed controller misuse of latent_mu #28

#30 AlexGonRo opened 5 years ago
0
Update README.md

#29 jilljenn closed 5 years ago
2
Possible error when predicting next action (class RolloutGenerator)

#28 AlexGonRo opened 5 years ago
7
one question about gmm_loss function

#27 UPUPGOO closed 4 years ago
1
Splitting of Train and Validation / Test set

#26 nlune closed 5 years ago
1
Error training MD-rnn

#25 waiyc opened 5 years ago
3
inconsisent MDRNN / MDRNNCell behavoir

#24 ranahanocka closed 5 years ago
3
sleep(0.1) leads to infty loops

#23 JiahaoYao closed 5 years ago
3
Moved "cur_best=None" outside MDRNN training loop

#22 Fairasname closed 5 years ago
2
Different transform in trainvae.py & trainmdrnn.py

#21 AmazingAng closed 5 years ago
2
Data generation script: No module named 'utils'

#20 0xsamgreen closed 5 years ago
3
Shouldn't this be outside the for loop?

#19 Fairasname closed 5 years ago
3
Negative GMM loss. How to interpret?

#18 Fairasname closed 5 years ago
2
[Question] Some controller training questions

#17 Fairasname closed 5 years ago
2
trainmdrnn running only test, the test loss decreases?

#16 ranahanocka closed 5 years ago
1
No init_hidden for mdrnn?

#15 czxttkl closed 5 years ago
2
[Error] Can not train VAE

#14 Fairasname closed 5 years ago
6
Reparameterization trick coefficient

#13 jordan-schneider closed 5 years ago
1
Does MDRNN loss include reward?

#12 mongoose54 closed 5 years ago
4
I try to train vae in pytorch 1.0, Failed

#11 eece-23 closed 5 years ago
3
Wrong dimension for pi softmax

#10 RobinKa closed 6 years ago
1
Controller training random stop

#9 wensdong closed 6 years ago
4
Fix incorrect help text for epochs

#8 superMDguy closed 6 years ago
0
untrained RNN

#7 wensdong closed 6 years ago
6
traincontroller issue - Stuck waiting for r_queue

#6 Narsil closed 5 years ago
10
Invalid argument array size

#5 Narsil closed 6 years ago
3
Update README.md

#4 Narsil closed 6 years ago
0
requirements are wrong ?

#3 Narsil closed 6 years ago
2
Having trouble generation_script.py

#2 whikwon closed 5 years ago
11
Fixes on few typos in the README

#1 diviyank closed 6 years ago
0