forked from fredzzhang/upt
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathlaunch_template.sh
51 lines (43 loc) · 2.92 KB
/
launch_template.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
#!/bin/bash
# This is a template for launching training and testing scripts
#
# Fred Zhang <[email protected]>
#
# The Australian National University
# Australian Centre for Robotic Vision
# -------------------------------
# Training commands
# -------------------------------
# Train UPT-R50 on HICO-DET
python main.py --world-size 8 --pretrained checkpoints/detr-r50-hicodet.pth --output-dir checkpoints/upt-r50-hicodet
# Train UPT-R101 on HICO-DET
python main.py --world-size 8 --backbone resnet101 --pretrained checkpoints/detr-r101-hicodet.pth --output-dir checkpoints/upt-r101-hicodet
# Train UPT-R101-DC5 on HICO-DET
python main.py --world-size 8 --backbone resnet101 --dilation --pretrained checkpoints/detr-r101-dc5-hicodet.pth --output-dir checkpoints/upt-r101-dc5-hicodet
# Train UPT-R50 on V-COCO
python main.py --world-size 8 --dataset vcoco --data-root vcoco/ --partitions trainval test --pretrained checkpoints/detr-r50-vcoco.pth --output-dir checkpoints/upt-r50-vcoco
# Train UPT-R101 on V-COCO
python main.py --world-size 8 --dataset vcoco --data-root vcoco/ --partitions trainval test --backbone resnet101 --pretrained checkpoints/detr-r101-vcoco.pth --output-dir checkpoints/upt-r101-vcoco
# Train UPT-R101-DC5 on V-COCO
python main.py --world-size 8 --dataset vcoco --data-root vcoco/ --partitions trainval test --backbone resnet101 --dilation --pretrained checkpoints/detr-r101-dc5-vcoco.pth --output-dir checkpoints/upt-r101-dc5-vcoco
# -------------------------------
# Testing commands
# -------------------------------
# Test UPT-R50 on HICO-DET
python main.py --eval --resume /path/to/model
# Test UPT-R101 on HICO-DET
python main.py --eval --backbone resnet101 --resume /path/to/model
# Test UPT-R101-DC5 on HICO-DET
python main.py --eval --backbone resnet101 --dilation --resume /path/to/model
# Cache detection results from UPT-R50 for Matlab evaluation on HICO-DET
python main.py --cache --output-dir matlab-r50 --resume /path/to/model
# Cache detection results from UPT-R101 for Matlab evaluation on HICO-DET
python main.py --cache --backbone resnet101 --output-dir matlab-r101 --resume /path/to/model
# Cache detection results from UPT-R101-DC5 for Matlab evaluation on HICO-DET
python main.py --cache --backbone resnet101 --dilation --output-dir matlab-r101-dc5 --resume /path/to/model
# Cache detection results from UPT-R50 for evaluation on V-COCO
python main.py --cache --dataset vcoco --data-root vcoco/ --partitions trainval test --output-dir vcoco-r50 --resume /path/to/model
# Cache detection results from UPT-R101 for evaluation on V-COCO
python main.py --cache --dataset vcoco --data-root vcoco/ --partitions trainval test --backbone resnet101 --output-dir vcoco-r101 --resume /path/to/model
# Cache detection results from UPT-R101-DC5 for evaluation on V-COCO
python main.py --cache --dataset vcoco --data-root vcoco/ --partitions trainval test --backbone resnet101 --dilation --output-dir vcoco-r101-dc5 --resume /path/to/model