issues
search
YeWR
/
EfficientZero
Open-source codebase for EfficientZero, from "Mastering Atari Games with Limited Data" at NeurIPS 2021.
GNU General Public License v3.0
839
stars
131
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
EfficientZero V2
#47
Michelle-Zou
opened
1 month ago
1
Fix a typo
#46
jackfirth
opened
6 months ago
0
Code for continuous action space
#45
davidva1
opened
1 year ago
1
ray warning
#44
QiGuLongDongQiang
opened
1 year ago
1
Question about the test phase not always running fully
#43
QiGuLongDongQiang
opened
1 year ago
1
Question about the effect of discount factor and done mask when calculating the target value?
#42
puyuan1996
opened
1 year ago
0
How to use with SLURM
#41
dillonmsandhu
opened
1 year ago
0
WSL2 NVIDIA 3090 or M1 MBP correct environment
#40
atalapan
opened
1 year ago
0
Question about whether need to train multiple agents for different games
#39
QiGuLongDongQiang
opened
1 year ago
1
In reanalyze_worker GPU worker, why prepare policy targets and value targets separately?
#38
desaixie
closed
1 year ago
2
Question about getting zero test score when I try to run EfficientZero on BabyAI grid environment
#37
jiachengc
opened
1 year ago
2
Question about the effect of state encoding indentity connection in dynamics network
#36
puyuan1996
closed
1 year ago
0
Question about the effect of state encoding indentity connection in dynamics network
#35
puyuan1996
opened
1 year ago
1
Question about the index of pad_child_visits_lst in selfplay_worker.py
#34
puyuan1996
opened
1 year ago
2
Question about the transform between true reward and value prefix
#33
timothijoe
opened
1 year ago
1
Cannot reproduce Breakout results
#32
vladisai
opened
1 year ago
0
Envs seem not to work in parallel
#31
AgentEXPL
opened
1 year ago
0
Add .gitignore for built ctree files
#30
steventrouble
closed
1 year ago
3
Clarification on the atari environment?
#29
alcinos
opened
2 years ago
0
reproduce results for other environment
#28
yix081
closed
2 years ago
0
The first selfplay worker uses the same seed for all parallel environments
#27
rPortelas
opened
2 years ago
2
EfficientZero high memory consumption / keeps increasing after replay buffer is full
#26
rPortelas
opened
2 years ago
7
procgen
#25
hlsafin
closed
2 years ago
2
Question about the effect of BatchNorm?
#24
jiaruonan
opened
2 years ago
0
Zero score on Freeway
#23
emailweixu
opened
2 years ago
6
How to evaluate the model
#22
yueyang130
closed
2 years ago
1
reproduce the result of CrazyClimber
#21
yueyang130
opened
2 years ago
13
Using custom gym environment
#20
PeterPirog
opened
2 years ago
0
Question about the effect of torch_amp
#19
puyuan1996
opened
2 years ago
2
Question about state_norm config option not mentioned in paper
#18
evanatyourservice
opened
2 years ago
0
Question about the dynamics network
#17
henrycharlesworth
opened
2 years ago
2
100% Offline RL use-case
#16
tbskrpmnns
opened
2 years ago
0
Reward clipping and value transformation
#15
zhixuan-lin
opened
2 years ago
1
Question: Why not reanalyze 100% policy targets?
#14
Hwhitetooth
opened
2 years ago
1
Slight discrepancy with implementation of value scaling
#13
henrycharlesworth
closed
2 years ago
2
Training is really slow
#12
SergioArnaud
opened
2 years ago
1
what does reward_hidden_c mean in mcts.py?
#11
sekv
closed
2 years ago
2
EfficientZero doesn't seem to be training
#10
SergioArnaud
closed
2 years ago
2
All memory seems on the first GPU
#9
geekyutao
opened
2 years ago
9
"bash make.sh" failed
#8
geekyutao
closed
2 years ago
3
BreakoutNoFrameskip-v4 is not registered?
#7
SpotEvictionPrediction
closed
2 years ago
2
License?
#6
charlesrwest
closed
2 years ago
1
Parametrize object store memory
#5
jl1990
closed
2 years ago
0
PyTorch Lightning Support
#4
tchaton
closed
1 year ago
2
Removing Baselines dependency
#3
araffin
closed
2 years ago
1
Fixed build errors.
#2
ipsec
opened
2 years ago
0
Where's the code?
#1
H-Park
closed
2 years ago
2