-
Notifications
You must be signed in to change notification settings - Fork 0
/
tune_online_rl.sh
42 lines (36 loc) · 1.58 KB
/
tune_online_rl.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
#!/bin/sh
start_day=7
n_train_days=1
n_windows=4 #8
n_seeds=3
n_trials=50
agent="DQN" #PPO, DQN, A2C
for ((i=3; i<n_windows; i++)); do
python3 tune_online_rl.py --start_day_train $((start_day + i)) \
--end_day_train $((start_day + i + n_train_days - 1)) \
--start_day_val $((start_day + i + n_train_days)) \
--end_day_val $((start_day + i + n_train_days)) \
--n_seeds $n_seeds \
--n_trials $n_trials \
--agent $agent \
--progress \
--out_dir "experiments/tuning/${agent}_window_${(start_day + i)}_${(start_day + i + n_train_days - 1)}" &
done
wait # Wait for all background processes to complete
#!/bin/sh
start_day_train=9
end_day_train=15
start_day_val=16
end_day_val=16
n_seeds=5
n_trials=50
agent="PPO" #PPO, DQN, A2C
python3 tune_online_rl.py --start_day_train $start_day_train \
--end_day_train $end_day_train \
--start_day_val $start_day_val \
--end_day_val $end_day_val \
--n_seeds $n_seeds \
--n_trials $n_trials \
--agent $agent \
--progress \
--out_dir "experiments/tuning/${agent}_${start_day_train}-${end_day_train}_${start_day_val}-${end_day_val}"