issues
search
nikhilbarhate99
/
PPO-PyTorch
Minimal implementation of clipped objective Proximal Policy Optimization (PPO) in PyTorch
MIT License
1.57k
stars
332
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
When to Update
#21
xunzhang
closed
4 years ago
5
PPO instead of PPO-M
#20
murtazabasu
closed
4 years ago
2
Shared parameters for NN action_layer and NN value_layer
#19
ArnoudWellens
closed
4 years ago
3
Ratio Calculation
#18
murtazabasu
closed
4 years ago
1
Minor change
#17
noanabeshima
closed
4 years ago
0
Ram gets full which stops the training session
#16
murtazabasu
closed
4 years ago
2
Learning from scratch without using pre-trained model
#15
EnnaSachdeva
closed
4 years ago
4
Why maintain two policies?
#14
biggzlar
closed
4 years ago
3
Question about GAE
#13
CatIIIIIIII
closed
4 years ago
1
Why are ratios not always 1?
#12
BigBadBurrow
closed
4 years ago
2
resetting timestep wrong?
#11
YilunZhou
closed
4 years ago
1
forget to copy policy to policy_old during ppo initialization?
#10
YilunZhou
closed
4 years ago
1
Generalized Advantage Estimation / GAE ?
#9
BigBadBurrow
closed
4 years ago
3
update() retains discounted_reward from previous episodes
#8
BigBadBurrow
closed
4 years ago
2
update() function to minimize, rather than maximize?
#7
BigBadBurrow
closed
4 years ago
2
Implementation issues
#6
kierkegaard13
closed
5 years ago
2
The `ppo_continuous.py` model does not learn
#5
chingandy
closed
5 years ago
2
PPO for continuous env
#4
zbenic
closed
5 years ago
1
how did you figure out continuous?
#3
nyck33
closed
5 years ago
1
edit advantage in surrogate
#2
AlpoGIT
closed
5 years ago
1
Create LICENSE
#1
nikhilbarhate99
closed
5 years ago
0
Previous