-
Notifications
You must be signed in to change notification settings - Fork 14
/
finetune_ret.sh
111 lines (83 loc) · 3.37 KB
/
finetune_ret.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
basedir=$1
# msrvtt ret
CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 --master_port 32711 ./train.py \
--pretrain_dir $basedir \
--config ./config/fast-retrieval-msrvtt.json \
--output_dir $basedir'/ret-msrvtt-lr2e-5-bs64-epoch5' \
--learning_rate 2e-5 \
--train_video_sample_num 4 \
--test_video_sample_num 8 \
--save_best true \
# msvd ret
# CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 --master_port 32711 ./train.py \
# --pretrain_dir $basedir \
# --config ./config/fast-retrieval-msvd.json \
# --output_dir $basedir'/ret-msvd-lr2e-5-bs64' \
# --learning_rate 2e-5 \
# --train_video_sample_num 4 \
# --test_video_sample_num 8 \
# --save_best true \
# activitynet ret
# CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 --master_port 32711 ./train.py \
# --pretrain_dir $basedir \
# --config ./config/fast-retrieval-activitynet.json \
# --output_dir $basedir'/ret-activitynet-lr2e-5-bs64-epoch40-frame8-audioframe4' \
# --learning_rate 2e-5 \
# --train_video_sample_num 8 \
# --train_audio_sample_num 4 \
# --test_video_sample_num 32 \
# --test_audio_sample_num 8 \
# --checkpointing true \
# --save_best true \
# valor32k ret
# CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 --master_port 32721 ./train.py \
# --pretrain_dir $basedir \
# --config ./config/fast-retrieval-valor32k.json \
# --output_dir $basedir'/ret-valor32k-lr2e-5-bs64-tva' \
# --learning_rate 2e-5 \
# --train_video_sample_num 4 \
# --test_video_sample_num 8 \
# --train_task 'ret%tva' \
# --test_task 'ret%tva' \
# didemo ret
# CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 --master_port 32711 ./train.py \
# --pretrain_dir $basedir \
# --config ./config/fast-retrieval-didemo.json \
# --output_dir $basedir'/ret-didemo-lr2e-5-bs64-epoch5-cliplr2.5e-6-lr9e-6' \
# --learning_rate 9e-6 \
# --train_video_sample_num 8 \
# --train_audio_sample_num 4 \
# --test_video_sample_num 32 \
# --test_audio_sample_num 8 \
# --train_epoch 30 \
# --save_best true \
# vatex ret
# CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 --master_port 32711 ./train.py \
# --pretrain_dir $basedir \
# --config ./config/fast-retrieval-vatex.json \
# --output_dir $basedir'/ret-vatex-50epoch-train4test8' \
# --learning_rate 2e-5 \
# --train_video_sample_num 4 \
# --test_video_sample_num 8 \
# --save_best true \
# lsmdc ret
# CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 --master_port 32711 ./train.py \
# --pretrain_dir $basedir \
# --config ./config/fast-retrieval-lsmdc.json \
# --output_dir $basedir'/ret-lsmdc-lr5e-7-bs64-epoch5-lr2e-5' \
# --learning_rate 2e-5 \
# --train_video_sample_num 4 \
# --test_video_sample_num 8 \
# --save_best true \
# --train_epoch 5 \
# mscoco ret res 392
# CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m torch.distributed.launch --nproc_per_node=8 --master_port 32711 ./train.py \
# --pretrain_dir $basedir \
# --config ./config/fast-retrieval-mscoco.json \
# --output_dir $basedir'/ret-mscoco-lr2e-5-bs1024-epoch15-res392' \
# --learning_rate 2e-5 \
# --video_resolution 392 \
# --train_batch_size 1024 \
# --train_epoch 75 \
# --checkpointing true \
# --save_best true \