jcwleo random-network-distillation-pytorch issues

jcwleo / random-network-distillation-pytorch

Random Network Distillation pytorch

MIT License

239 stars 43 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Extrinsic reward clipping

#35 cangozpi opened 6 months ago
0
Action values are incremented by 1 for the Breakout game ?

#34 cangozpi opened 6 months ago
0
Intrinsic reward calculation, sum or mean?

#33 aklein1995 opened 3 years ago
2
if i want to employe this work to a new env, what should i do?

#32 SOMEAIDI opened 3 years ago
0
training error

#31 rainbow979 opened 4 years ago
1
About sticky action

#30 tongzhoumu opened 5 years ago
1
input_size in model.py?

#29 chaityabshah closed 5 years ago
1
I tried to train system but get error

#28 rnunziata closed 5 years ago
1
what is meaning of line in envs.py:

#27 rnunziata closed 5 years ago
2
How long did you get 6100?

#26 zhr211 opened 5 years ago
6
Generalized Advantage Estimator problem

#25 RozenAstrayChen closed 5 years ago
2
Issue: applied mario env

#24 jcwleo closed 5 years ago
0
Hotfix: default conf as in paper

#23 kslazarev closed 5 years ago
0
Hotfix: visited rooms on done

#22 kslazarev closed 5 years ago
0
Hotfix: reset reward on done

#21 kslazarev closed 5 years ago
1
Reduce memory usage (2-3x times)

#20 kslazarev opened 5 years ago
0
Mario eval is slow

#19 simoninithomas closed 5 years ago
2
리워드 필터에 쓰이는 파라미터 수정

#18 jcwleo closed 5 years ago
0
global_grad_norm_ has no effect

#17 shuang-liu opened 5 years ago
2
Line 68 in train.py

#16 shuang-liu closed 5 years ago
1
Use several GPUs if they exist

#15 kslazarev closed 5 years ago
1
TypeError: can't convert CUDA tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first

#14 kslazarev closed 5 years ago
2
Input shape is not correct in Linear-8 layer in CnnActorCriticNetwork feature model

#13 kslazarev closed 5 years ago
1
모델 업데이트

#12 jcwleo closed 5 years ago
0
global_norm 추가

#11 jcwleo closed 5 years ago
0
Reward converge at 4600

#10 Acmece closed 5 years ago
1
Add logging information for Montezuma's Revenge

#9 jcwleo closed 5 years ago
0
RNN 모델 추가

#8 jcwleo opened 5 years ago
0
Issue/applied paper parameter

#7 jcwleo closed 5 years ago
0
refactoring make_train_data

#6 jcwleo closed 5 years ago
0
README asset

#5 jcwleo opened 5 years ago
18
논문에 나온 파라미터 적용

#4 jcwleo closed 5 years ago
0
rnd 모델에서는 1-stack을 사용

#3 jcwleo closed 5 years ago
0
fixed model network initialization

#2 jcwleo closed 5 years ago
0
add life done option

#1 jcwleo closed 5 years ago
0