pranz24 pytorch-soft-actor-critic issues

pranz24 / pytorch-soft-actor-critic

PyTorch implementation of soft actor critic

MIT License

810 stars 180 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Question: Why optimize loss_alpha?

#47 DefinitlyEvil opened 8 months ago
0
No normalization of state space

#46 rosa-wolf opened 9 months ago
0
Training policy for more complex tasks, converges to sub-optimal solutions

#45 rosa-wolf opened 9 months ago
0
the bound enforce for log_prob in line 103 of model.py

#44 Roboticyang opened 9 months ago
0
Is this code SAC-V not a SAC?

#43 night2570 closed 8 months ago
1
[Question] Mask Batch

#42 chenxi-yang closed 8 months ago
2
Model saving and loading

#41 tissten opened 2 years ago
1
question about q_loss and alpha_loss

#40 xxxkxin closed 2 years ago
0
Doubts about Regularization in policy loss

#39 Marxvans opened 2 years ago
0
fix module import

#38 Gloriel621 closed 2 years ago
0
Improved model saving with Checkpoint system

#37 shnippi closed 3 years ago
1
continuous cartpole trains

#36 jonberliner closed 8 months ago
2
Resume training

#35 Tomeu7 opened 3 years ago
5
Exploding entropy temperature

#34 reubenwong97 opened 4 years ago
10
multiplying action_scale in the log_prob computation

#33 sungsulim closed 4 years ago
1
Inconsistent seeding

#32 mohakbhardwaj closed 4 years ago
2
Running SAC: Operation failed to compute its gradient

#31 ian-cannon opened 4 years ago
3
Support OpenAI Gym Robotic Env?

#30 peiseng opened 4 years ago
0
what is the derivation behind the log_prob equation?

#29 FernandoCamaro closed 4 years ago
1
Update main.py

#28 llucid-97 closed 4 years ago
1
Action scaling is missing on SAC_V branch

#27 alirezakazemipour closed 4 years ago
4
Can I use this in custom gym env?

#26 kwk2696 opened 4 years ago
1
Target value calculation maistake

#25 alirezakazemipour closed 4 years ago
4
Action scale and action bias

#24 shakenov-chinga opened 4 years ago
1
does your code support multi-dimension discrete action space?

#23 KK666-AI closed 4 years ago
1
Why is the value function not used in this implementation?

#22 Steven-Ho closed 4 years ago
4
Fix error with DeterministicPolicy

#21 Shmuma closed 4 years ago
1
puzzles about action scaling

#20 wayunderfoot closed 4 years ago
2
Update README.md

#19 fgolemo closed 5 years ago
1
Could you please explain the "# Enforcing Action Bound" comment?

#18 Wen-Wen-Luffy closed 5 years ago
3
About model.py line 105

#17 BangLiu closed 5 years ago
3
Unable to reproduce results on Humanoid-v2 in new SAC

#16 zwfightzw opened 5 years ago
6
Fix bugs of action re-scaling

#15 toshikwa closed 5 years ago
5
Question about policy_loss

#14 toshikwa closed 5 years ago
11
Fix normalized actions

#13 toshikwa closed 5 years ago
1
Normalized Actions has bugs

#12 Phlogiston90 closed 5 years ago
3
Derivative in reparametrization trick?

#11 ZeratuuLL closed 5 years ago
9
A little question about calculating log likelihood

#10 ZeratuuLL closed 5 years ago
1
Value network

#9 jendelel closed 5 years ago
4
Cuda support

#8 jendelel closed 5 years ago
2
Fixed bug with double tensoring the initial state.

#7 jendelel closed 5 years ago
1
Why do you need to use NormalizedActions()?

#6 JingJerry closed 5 years ago
4
reparametrization trick issue

#5 tldoan closed 5 years ago
10
reproducibility for HalfCheetah-v2

#4 tldoan closed 5 years ago
5
Policy Loss with Minimum or Q1?

#3 pranv closed 5 years ago
4
A question in the deterministic case

#2 roosephu closed 5 years ago
3
Images

#1 pranz24 closed 6 years ago
1