/
script_plots.sh
59 lines (37 loc) · 2.73 KB
/
script_plots.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
## To create Fig. 1 and Fig. 3
declare -a ModelTypes=('resnet' 'fcnet' 'vit' 'densenet' 'vgg' 'MLPMixer4')
for mod in "${ModelTypes[@]}";
do
for seed in 0 1 2
do
python main.py --net $mod --load_net <path_to_saved_models>.pth --plot_path ./images/paper/ --imgs 30,72,42 --set_seed $seed --range_l 0.5 --range_r 0.5
done
done
for wid in 10 20 30
do
for seed in 0 1 2
do
python main.py --net WideResNet --widen_factor $wid --load_net <path_to_saved_models>.pth --plot_path ./images/paper/ --imgs 30,72,42 --set_seed $seed --range_l 0.5 --range_r 0.5
done
done
### Plot uniform noise and random shuffled images (i.e. Figs. 2, 11, 12)
declare -a ModelTypes=('resnet' 'fcnet' 'vit' 'densenet' 'vgg' 'MLPMixer4')
for mod in "${ModelTypes[@]}";
do
for seed in 0 1 2 3 4 5 6 7 8 9
do
python main.py --net $mod --noise_type uniform_random --plot_path ./images/paper/uniform_noise/ --load_net <path_to_saved_models>.pth --imgs 10,20,30 --set_seed $seed --range_l 0.5 --range_r 0.5
python main.py --net $mod --noise_type random_shuffle --plot_path ./images/paper/random_shuffle/ --load_net <path_to_saved_models>.pth --imgs 10,20,30 --set_seed $seed --range_l 0.5 --range_r 0.5
done
done
### Mixup plots (Fig. 13)
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/mixup --imgs 3,84,18 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/mixup --imgs 19,22,72 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/mixup --imgs 46,43,67 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/mixup --imgs 62,27,3 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/mixup --imgs 111,71,115 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/nomixup --imgs 3,84,18 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/nomixup --imgs 19,22,72 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/nomixup --imgs 46,43,67 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/nomixup --imgs 62,27,3 --set_seed 0 --range_l 0.5 --range_r 0.5
python main.py --net resnet --load_net <path_to_saved_models>.pth --plot_path ./images/paper/nomixup --imgs 111,71,115 --set_seed 0 --range_l 0.5 --range_r 0.5