-
Notifications
You must be signed in to change notification settings - Fork 1
/
esub.sh
executable file
·138 lines (94 loc) · 5.7 KB
/
esub.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
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
#!/bin/bash
DATE=$(date +%Y_%m_%d_%H%M)
OUTPUT_DIR=$(printf "%s/%s/%s" $SCRATCH "DL" $DATE)
mkdir -p $OUTPUT_DIR
ITERATIONS=5
TRAJECTORIES=500
# for game in lunar_lander cart_pole; do
# for estimator in Reinforce Gpomdp SarahPg Svrpg StormPg PagePg Svrpg_auto PageStormPg PagePg; do
# output=$(printf "%s_%s_:output.txt" $game $estimator)
# echo $output
# bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES
# done
# done
# # bsub -W 24:00 -n 2 -R "rusage[mem=4096]" -J "cart_gpomdp" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES
for game in cart_pole lunar_lander; do
for estimator in Gpomdp SarahPg Svrpg StormPg PagePg PageStormPg; do
output=$(printf "%s_%s_bs:%s_mbs:%s_alpha:%s_lr:%s_prob:%s.txt" $game $estimator $batch_size $mini_batch_size $alpha $lr $prob)
echo $output
bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python main.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES
done
done
# for game in lunar_lander; do
# for estimator in Gpomdp SarahPg Svrpg StormPg PagePg PageStormPg; do
# for batch_size in 10 25 100; do
# for mini_batch_size in 5 10; do
# for alpha in 0.7 0.75 0.8 0.85 0.9 0.95; do
# for lr in 0.001 0.005 0.01 0.05; do
# for prob in 0.7 0.75 0.8 0.85 0.9; do
# output=$(printf "%s_%s_bs:%s_mbs:%s_alpha:%s_lr:%s_prob:%s.txt" $game $estimator $batch_size $mini_batch_size $alpha $lr $prob)
# echo $output
# bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES --prob $prob --alpha $alpha --batch_size $batch_size --mini_batch_size $mini_batch_size --lr $lr
# done
# done
# done
# done
# done
# done
# done
# for game in cart_pole lunar_lander mountain_car continuous_mountain_car; do
# for estimator in PageStormPg; do
# for probability in $(seq 0.1 0.1 0.9); do
# output=$(printf "%s_%s_prob:%s_output.txt" $game $estimator $probability)
# echo $output
# bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES --prob $probability
# done
# done
# for estimator in PageStormPg StormPg; do
# for alpha in $(seq 0 0.1 1); do
# output=$(printf "%s_%s_alpha:%s_output.txt" $game $estimator $alpha)
# echo $output
# bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES --alpha $alpha
# done
# done
# for estimator in Reinforce Gpomdp SarahPg Svrpg StormPg PagePg Svrpg_auto PageStormPg PagePg; do
# lr_s=( 1e-3 5e-3 1e-2 5e-2 )
# flr_s=( 2e-3 1e-2 2e-2 1e-1 )
# for ((i=0;i<${#lr_s[@]};i++))
# do
# lr=${lr_s[$i]}
# flr=${flr_s[$i]}
# output=$(printf "%s_%s_lr%sx%s:output.txt" $game $estimator $lr $flr)
# echo $output
# bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES --lr $lr --flr $flr
# done
# done
# done
# for game in cart_pole lunar_lander mountain_car continuous_mountain_car; do
# for estimator in PageStormPg PagePg; do
# for probability in $(seq 0.1 0.1 0.9); do
# output=$(printf "%s_%s_prob:%s_output.txt" $game $estimator $probability)
# echo $output
# bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES --prob $probability
# done
# done
# for estimator in PageStormPg StormPg; do
# for alpha in $(seq 0 0.1 1); do
# output=$(printf "%s_%s_alpha:%s_output.txt" $game $estimator $alpha)
# echo $output
# bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES --alpha $alpha
# done
# done
# for estimator in Reinforce Gpomdp SarahPg Svrpg StormPg PagePg Svrpg_auto PageStormPg PagePg; do
# lr_s=( 1e-3 5e-3 1e-2 5e-2 )
# flr_s=( 2e-3 1e-2 2e-2 1e-1 )
# for ((i=0;i<${#lr_s[@]};i++))
# do
# lr=${lr_s[$i]}
# flr=${flr_s[$i]}
# output=$(printf "%s_%s_lr%sx%s:output.txt" $game $estimator $lr $flr)
# echo $output
# bsub -W 24:00 -n 1 -R "rusage[mem=4096]" -J "$game" -oo $OUTPUT_DIR/$output python environment.py --game $game --estimator $estimator --output $OUTPUT_DIR --iter $ITERATIONS --num_traj $TRAJECTORIES --lr $lr --flr $flr
# done
# done
# done