forked from swagshaw/Rainbow-Keywords
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathaug_experiment.sh
102 lines (92 loc) · 4.25 KB
/
aug_experiment.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
#/bin/bash
# CIL CONFIG
MODE="rainbow_keywords" # finetune ,native_rehearsal,joint, rwalk, icarl, rainbow_keywords, ewc, bic, gdumb
# "default": If you want to use the default memory management method.
MEM_MANAGE="default" # default, random, reservoir, uncertainty, prototype.
RND_SEED=1
DATASET="gsc" # gsc
STREAM="online" # offline, online
EXP="disjoint" # disjoint, blurry10, blurry30
MEM_SIZE=500 # k={200, 500, 1000}
TRANS="" # multiple choices:"Additional train transforms [mixup, specmix, specaugment]"
TOPK=1
N_WORKER=5
JOINT_ACC=0.94 # training all the tasks at once.
N_TASKS=6
N_INIT_CLS=15
N_CLS_A_TASK=3
MODEL_NAME="tcresnet8"
OPT_NAME="adam"
SCHED_NAME="cos"
LR=0.1
BATCHSIZE=128
N_EPOCH=50
# FINISH CIL CONFIG ####################
UNCERT_METRIC="vr"
PRETRAIN="" INIT_MODEL="" INIT_OPT="--init_opt"
# for iCaRL
FEAT_SIZE=256
# for BiC
distilling="" # Normal BiC. If you do not want to use distilling loss, then "".
python main.py --mode efficient_memory --mem_manage $MEM_MANAGE --exp_name $EXP \
--dataset $DATASET \
--stream_env $STREAM $INIT_MODEL $INIT_OPT --topk $TOPK \
--n_tasks $N_TASKS --n_cls_a_task $N_CLS_A_TASK --n_init_cls $N_INIT_CLS \
--rnd_seed $RND_SEED \
--model_name $MODEL_NAME --opt_name $OPT_NAME $PRETRAIN --sched_name $SCHED_NAME \
--lr $LR --batchsize $BATCHSIZE \
--n_worker $N_WORKER --n_epoch $N_EPOCH \
--memory_size 300 --transform "kd_trick" --uncert_metric $UNCERT_METRIC \
--feature_size $FEAT_SIZE $distilling --joint_acc $JOINT_ACC
python main.py --mode efficient_memory --mem_manage $MEM_MANAGE --exp_name $EXP \
--dataset $DATASET \
--stream_env $STREAM $INIT_MODEL $INIT_OPT --topk $TOPK \
--n_tasks $N_TASKS --n_cls_a_task $N_CLS_A_TASK --n_init_cls $N_INIT_CLS \
--rnd_seed $RND_SEED \
--model_name $MODEL_NAME --opt_name $OPT_NAME $PRETRAIN --sched_name $SCHED_NAME \
--lr $LR --batchsize $BATCHSIZE \
--n_worker $N_WORKER --n_epoch $N_EPOCH \
--memory_size 1500 --transform "kd_trick" --uncert_metric $UNCERT_METRIC \
--feature_size $FEAT_SIZE $distilling --joint_acc $JOINT_ACC
python main.py --mode efficient_memory --mem_manage $MEM_MANAGE --exp_name $EXP \
--dataset $DATASET \
--stream_env $STREAM $INIT_MODEL $INIT_OPT --topk $TOPK \
--n_tasks $N_TASKS --n_cls_a_task $N_CLS_A_TASK --n_init_cls $N_INIT_CLS \
--rnd_seed $RND_SEED \
--model_name $MODEL_NAME --opt_name $OPT_NAME $PRETRAIN --sched_name $SCHED_NAME \
--lr $LR --batchsize $BATCHSIZE \
--n_worker $N_WORKER --n_epoch $N_EPOCH \
--memory_size 1000 --transform "kd_trick" --uncert_metric $UNCERT_METRIC \
--feature_size $FEAT_SIZE $distilling --joint_acc $JOINT_ACC
#python main.py --mode efficient_memory --mem_manage $MEM_MANAGE --exp_name $EXP \
#--dataset $DATASET \
#--stream_env $STREAM $INIT_MODEL $INIT_OPT --topk $TOPK \
#--n_tasks $N_TASKS --n_cls_a_task $N_CLS_A_TASK --n_init_cls $N_INIT_CLS \
#--rnd_seed $RND_SEED \
#--model_name $MODEL_NAME --opt_name $OPT_NAME $PRETRAIN --sched_name $SCHED_NAME \
#--lr $LR --batchsize $BATCHSIZE \
#--n_worker $N_WORKER --n_epoch $N_EPOCH \
#--memory_size 500 --transform "kd_trick review specaug labels_trick" --uncert_metric $UNCERT_METRIC \
#--feature_size $FEAT_SIZE $distilling --joint_acc $JOINT_ACC
#
#
#python main.py --mode efficient_memory --mem_manage $MEM_MANAGE --exp_name $EXP \
#--dataset $DATASET \
#--stream_env $STREAM $INIT_MODEL $INIT_OPT --topk $TOPK \
#--n_tasks $N_TASKS --n_cls_a_task $N_CLS_A_TASK --n_init_cls $N_INIT_CLS \
#--rnd_seed $RND_SEED \
#--model_name $MODEL_NAME --opt_name $OPT_NAME $PRETRAIN --sched_name $SCHED_NAME \
#--lr $LR --batchsize $BATCHSIZE \
#--n_worker $N_WORKER --n_epoch $N_EPOCH \
#--memory_size 500 --transform "kd_trick mixup review specaug labels_trick" --uncert_metric $UNCERT_METRIC \
#--feature_size $FEAT_SIZE $distilling --joint_acc $JOINT_ACC
#python main.py --mode efficient_memory --mem_manage $MEM_MANAGE --exp_name $EXP \
#--dataset $DATASET \
#--stream_env $STREAM $INIT_MODEL $INIT_OPT --topk $TOPK \
#--n_tasks $N_TASKS --n_cls_a_task $N_CLS_A_TASK --n_init_cls $N_INIT_CLS \
#--rnd_seed $RND_SEED \
#--model_name $MODEL_NAME --opt_name $OPT_NAME $PRETRAIN --sched_name $SCHED_NAME \
#--lr $LR --batchsize $BATCHSIZE \
#--n_worker $N_WORKER --n_epoch $N_EPOCH \
#--memory_size 500 --transform "review" --uncert_metric $UNCERT_METRIC \
#--feature_size $FEAT_SIZE $distilling --joint_acc $JOINT_ACC