tambetm simple_dqn issues

tambetm / simple_dqn

Simple deep Q-learning agent.

MIT License

692 stars 184 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Neon installation

#57 jboilard1994 closed 5 years ago
3
snapshots saving as prm

#56 andynguyen7 closed 6 years ago
1
README.md: for Ubuntu cv2.so changed to cv2.x86_64-linux-gnu.so

#55 tgianko opened 6 years ago
1
ValueError: backend must be one of ('cpu', 'mkl')

#54 bin913 closed 6 years ago
1
test with gym

#53 mcbrs1a closed 6 years ago
1
Assertion Error

#52 mcbrs1a opened 6 years ago
1
when running /play.sh on cpu

#51 mcbrs1a closed 6 years ago
1
RuntimeError: Device 0 does not have CUDA compute capability 3.0 or greater

#50 riccitensor closed 6 years ago
3
slow training speed in latest code?

#49 mw66 opened 6 years ago
9
Different absorbing states in ale and gym

#48 carloderamo closed 7 years ago
1
1.Fix play_random in which env restart must be called before step.2.Fix readme

#47 cxxgtxy closed 7 years ago
2
Gym reset must be called before step or "AssertionError: Cannot call …

#46 cxxgtxy closed 7 years ago
1
AssertionError: Cannot call env.step() before calling reset(). How to remove the error?

#45 ayushya2531 closed 7 years ago
2
What's the purpose of _restartRandom?

#44 gauss-clb closed 7 years ago
5
error pi3 for test_gym.py and main.py,

#43 Robomate closed 7 years ago
1
error on PI3 for ./play.sh snapshots/breakout_77.pkl

#42 Robomate closed 7 years ago
7
SyntaxError

#41 Dahandla closed 7 years ago
1
simple_dqn code test problems

#40 ctampier closed 7 years ago
7
Avoid a life loss from stopping the training process - Attempt 2

#39 ferminquant closed 7 years ago
3
Avoid a life loss from stopping the training process

#38 ferminquant closed 7 years ago
2
Error when training: "terminal state occurred during random initialization"

#37 ferminquant closed 7 years ago
5
Update README to accomodate recent change

#36 mthrok closed 7 years ago
0
Treat life loss as termination in training mode

#35 mthrok closed 7 years ago
2
Change network sync timing from #trainings to #steps

#34 mthrok closed 7 years ago
1
Change parameter initialization to Xavier

#33 mthrok closed 8 years ago
0
The test score is different from the DeepMind paper

#32 futurecrew opened 8 years ago
14
Make play_random to fill memory

#31 mthrok closed 8 years ago
1
targets are not really copied when using cpu backend

#30 loofahcus closed 8 years ago
1
I want to make my own ale models for DQN Reinforce Learning using my video file input.

#29 bemoregt closed 8 years ago
1
error when use openAI gym

#28 nina124 closed 7 years ago
2
AttributeError: 'numpy.ndarray' object has no attribute 'asnumpyarray'

#27 Seraphli closed 8 years ago
1
in neon new release v1.5.1: cost.asnumpyarray() error out

#26 mw66 closed 8 years ago
0
src/environment.py: (args.screen_width, args.screen_height) wrong order

#25 mw66 closed 8 years ago
2
self.model.initialize(self.input_shape[:-1], self.cost)? correct dataset param?

#24 mw66 closed 8 years ago
2
./play.sh snapshots/space_invaders_200.pkl error out

#23 mw66 opened 8 years ago
2
What's the reason to add StateBuffer?

#22 mw66 closed 8 years ago
9
Array is too big when I try to begin training

#21 ashwinreddy closed 8 years ago
1
./test_gym.sh 'NoneType' object has no attribute 'add'

#20 iaroslav-ai closed 8 years ago
2
./test_gym.sh error

#19 iaroslav-ai closed 8 years ago
0
can the train process with CPU backend utilize multi-cores?

#18 loofahcus closed 8 years ago
2
memory error

#17 FadeDreams closed 8 years ago
1
ImportError: No module named cv2

#16 FadeDreams closed 8 years ago
6
Visual and Test not work ?

#15 mingtop closed 8 years ago
2
Memory Error

#14 ShawnWongMILab closed 8 years ago
3
resume training will erase old results/game.csv & overwrite snapshots/game_$(EPOCH).pkl from 1

#13 mw66 closed 8 years ago
1
src/nvis/figure.py upgrade to 0.11.1?

#12 mw66 closed 8 years ago
1
should _restartRandom (src/agent.py) choose random action?

#11 mw66 closed 8 years ago
2
small simplification: src/replay_memory.py line 35, 39

#10 mw66 closed 8 years ago
1
nvis.sh: pycuda._driver.LogicError: cuFuncSetBlockShape failed: an illegal memory access was encountered

#9 mw66 closed 8 years ago
3
Performance Update

#8 jcoreyes opened 8 years ago
2