Hw1/learn#1

Open

CeHao1 wants to merge 36 commits intoberkeleydeeprlcourse:mainfrom

CeHao1:hw1/learn

Commits on Sep 1, 2021

1
ce-hao
committed
finish two utils
ce-hao
committed
finish the replay sample
ce-hao
committed
finish the replay sample
ce-hao
committed
checked two utils, they are correct;
ce-hao
committed
other students use strange ways in replay buffer
ce-hao
committed
ok trainer
ce-hao
committed
mlp policy get action has many implementations
ce-hao
committed
this seems ok
ce-hao
committed

Commits on Sep 3, 2021

add pickle
ce-hao
committed
add self.sample_recent_data
ce-hao
committed
change sample random
ce-hao
committed
change sample random
ce-hao
committed
add reset
ce-hao
committed
this tried to imitate human
ce-hao
committed
action
ce-hao
committed
action
ce-hao
committed
action
ce-hao
committed
1
ce-hao
committed
1
ce-hao
committed
1
ce-hao
committed
start the plot frame
ce-hao
committed
the plot is ok
ce-hao
committed
the plot is ok
ce-hao
committed
1
ce-hao
committed
1
ce-hao
committed
it can save
ce-hao
committed
1
ce-hao
committed
copy from colab
ce-hao
committed
start adding more ep_len
ce-hao
committed
to test the expert policy
ce-hao
committed
last forgot eval reward
ce-hao
committed
this one is good for ploting
ce-hao
committed
super cool;
ce-hao
committed
yes
ce-hao
committed
removed unuseful files
ce-hao
committed