-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathrun.sh
executable file
·98 lines (68 loc) · 8.18 KB
/
run.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
#!/usr/bin/env bash
BASEDIR=$(dirname "$0")
# echo "$BASEDIR"
# ------- training agent using TRPO algorithm -------
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=ttr --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=hand_craft --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_0.1 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_1 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_10 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=DubinsCarEnv-v0 --reward_type=ttr --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=DubinsCarEnv-v0 --reward_type=hand_craft --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance_lambda_0.1 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance_lambda_1 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance_lambda_10 --set_additional_goal=angle
# run 5 times for each reward type using trpo algorithm for quadrotor task
#for VARIABLE in 1 2 3 4 5
#do
# python3.5 $BASEDIR/train_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=ttr --algo=trpo --set_angle_goal=false
# python3.5 $BASEDIR/train_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=hand_craft --algo=trpo --set_angle_goal=false
# python3.5 $BASEDIR/train_trpo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance --algo=trpo --set_angle_goal=false
#done
# ---------------------------------------------------
# ------- training agent using PPO algorithm -------
# python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=ttr --algo=ppo --set_additional_goal=angle
#python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance --algo=ppo --set_additional_goal=angle
#python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=hand_craft --algo=ppo --set_additional_goal=angle
#python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_0.1 --algo=ppo --set_additional_goal=angle
##python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_1 --algo=ppo --set_additional_goal=angle
#python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_10 --algo=ppo --set_additional_goal=angle
#python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_0.1 --algo=ppo --set_additional_goal=angl
#python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_10 --algo=ppo --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=DubinsCarEnv-v0 --reward_type=ttr --algo=ppo --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=DubinsCarEnv-v0 --reward_type=hand_craft --algo=ppo --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance --algo=ppo --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance_lambda_0.1 --algo=ppo --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance_lambda_1 --algo=ppo --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance_lambda_10 --algo=ppo --set_additional_goal=angle
# run 5 times for each reward type using ppo algorithm for quadrotor task
#for VARIABLE in 1 2 3 4 5
#do
# python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=ttr --algo=ppo --set_angle_goal=false
# python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=hand_craft --algo=ppo --set_angle_goal=false
# python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance --algo=ppo --set_angle_goal=false
#done
# --------------------------------------------------
# ------- training agent using DDPG algorithm -------
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ddpg.py --env_id=PlanarQuadEnv-v0 --reward_type=ttr --set_additional_goal=angle
#python3.5 $BASEDIR/train_ddpg.py --env_id=PlanarQuadEnv-v0 --reward_type=distance --set_additional_goal=angle
#python3.5 $BASEDIR/train_ddpg.py --env_id=PlanarQuadEnv-v0 --reward_type=hand_craft --set_additional_goal=angle
#python3.5 $BASEDIR/train_ddpg.py --env_id=PlanarQuadEnv-v0 --reward_type=distance_lambda_0.1 --set_additional_goal=angle
#python3.5 $BASEDIR/train_ddpg.py --env_id=PlanarQuadEnv-v0 --reward_type=distance_lambda_1 --set_additional_goal=angle
#python3.5 $BASEDIR/train_ddpg.py --env_id=PlanarQuadEnv-v0 --reward_type=distance_lambda_10 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ddpg.py --env_id=DubinsCarEnv-v0 --reward_type=ttr --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ddpg.py --env_id=DubinsCarEnv-v0 --reward_type=hand_craft --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ddpg.py --env_id=DubinsCarEnv-v0 --reward_type=distance --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ddpg.py --env_id=DubinsCarEnv-v0 --reward_type=distance_lambda_0.1 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ddpg.py --env_id=DubinsCarEnv-v0 --reward_type=distance_lambda_1 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ddpg.py --env_id=DubinsCarEnv-v0 --reward_type=distance_lambda_10 --set_additional_goal=angle
# ---------------------------------------------------
# some extra experiments
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=DubinsCarEnv-v0 --reward_type=ttr --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/quick_trpo.py --gym_env=DubinsCarEnv-v0 --reward_type=distance_lambda_1 --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=DubinsCarEnv-v0 --reward_type=ttr --algo=ppo --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=ttr --algo=ppo --set_additional_goal=angle
#/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=PlanarQuadEnv-v0 --reward_type=distance_lambda_0.1 --algo=ppo --set_additional_goal=angle
/local-scratch/xlv/miniconda3/envs/py35_no_specific/bin/python3.5 $BASEDIR/train_ppo.py --gym_env=DubinsCarEnv-v0 --reward_type=ttr --algo=ppo --set_additional_goal=angle