-
Notifications
You must be signed in to change notification settings - Fork 4
/
run_all_mini-ImageNet-LT.sh
executable file
·71 lines (54 loc) · 2.43 KB
/
run_all_mini-ImageNet-LT.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
# mini-ImageNet-LT
# Table 4
#----------------------------------------------------------
# Baseline
# -[x] CE (0.1) # Pre-requisite: None
# -[x] CosineCE (0.2) # Pre-requisite: None
#----------------------------------------------------------
# Decouple
# -[x] cRT (0.3) # Pre-requisite: Experiment "0.1"
#----------------------------------------------------------
# Distillation
# -[x] CBD (0.4) # Pre-requisite: Experiment "0.2" but with seeds 10
# For repo specific to CBD paper with much more detailed instructions, check https://github.com/rahulvigneswaran/Class-Balanced-Distillation-for-Long-Tailed-Visual-Recognition.pytorch
#----------------------------------------------------------
# Generation
# -[x] MODALS (0.5) # Pre-requisite: Experiment "0.1"
#----------------------------------------------------------
# Ours
# -[x] CosineCE + TailCalib (1.2) # Pre-requisite: Experiment "0.2"
# -[x] CosineCE + TailCalibX (2.2) # Pre-requisite: Experiment "0.2"
#----------------------------------------------------------
#----mini-ImageNet_LT
actual_dataset=2
seeds=1
#----CE
experiment_no=0.1
python main.py --experiment=$experiment_no --gpu="1" --dataset=$actual_dataset --seed=$seeds --train
wait
# ----CosineCE
experiment_no=0.2
python main.py --experiment=$experiment_no --gpu="1" --dataset=$actual_dataset --seed=$seeds --train
python main.py --experiment=$experiment_no --gpu="1" --dataset=$actual_dataset --seed=10 --train #For CBD
wait
#----cRT
experiment_no=0.3
python main.py --experiment=$experiment_no --gpu="1" --dataset=$actual_dataset --seed=$seeds --train
wait
#----CBD
experiment_no=0.4
python main.py --experiment=$experiment_no --gpu="1" --dataset=$actual_dataset --seed=$seeds --cv1=0.4 --cv2=100 --train
wait
#----MODALS
experiment_no=0.5
python main.py --experiment=$experiment_no --gpu="1" --dataset=$actual_dataset --seed=$seeds --generate --cv1=0.01 --train
wait
#----CosineCE + TailCalib
experiment_no=1.2
python main.py --experiment=$experiment_no --gpu="1" --dataset=$actual_dataset --cv1=1.0 --cv2=0.01 --cv3=0.7 --cv4=0.0 --cv5=3 --generate --retraining
wait
#----CosineCE + TailCalibX
experiment_no=2.2
python main.py --experiment=$experiment_no --gpu="1" --dataset=$actual_dataset --cv1=1.0 --cv2=0.01 --cv3=0.7 --cv4=0.0 --cv5=3 --train
wait
#--------------------------------------------------- Paper experiments end here