-
Notifications
You must be signed in to change notification settings - Fork 183
/
cfgs_res50_dota2.0_kf_v4.py
85 lines (66 loc) · 2.17 KB
/
cfgs_res50_dota2.0_kf_v4.py
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
# -*- coding: utf-8 -*-
from __future__ import division, print_function, absolute_import
import numpy as np
from alpharotate.utils.pretrain_zoo import PretrainModelZoo
from configs._base_.models.retinanet_r50_fpn import *
from configs._base_.datasets.dota_detection import *
from configs._base_.schedules.schedule_1x import *
# schedule
BATCH_SIZE = 1
GPU_GROUP = "0"
NUM_GPU = len(GPU_GROUP.strip().split(','))
SAVE_WEIGHTS_INTE = 40000
DECAY_STEP = np.array(DECAY_EPOCH, np.int32) * SAVE_WEIGHTS_INTE
MAX_ITERATION = SAVE_WEIGHTS_INTE * MAX_EPOCH
WARM_SETP = int(WARM_EPOCH * SAVE_WEIGHTS_INTE)
# dataset
DATASET_NAME = 'DOTA2.0'
CLASS_NUM = 18
# model
# backbone
pretrain_zoo = PretrainModelZoo()
PRETRAINED_CKPT = pretrain_zoo.pretrain_weight_path(NET_NAME, ROOT_PATH)
TRAINED_CKPT = os.path.join(ROOT_PATH, 'output/trained_weights')
# loss
CENTER_LOSS_MODE = 1 # center loss in kld
CLS_WEIGHT = 1.0
REG_WEIGHT = 0.01
VERSION = 'RetinaNet_DOTA2.0_KF_1x_20220920'
"""
RetinaNet-H + log(kl_center) + kfiou (exp(1-IoU)-1)
loss = (loss_1.reshape([n, 1]) + loss_2).reshape([n*n,1])
loss = sum(loss)
loss /= n
FLOPs: 487527642; Trainable params: 33148131
This is your evaluation result for task 1 (VOC metrics):
mAP: 0.4894285127967532
ap of each class:
plane:0.7855100097732065,
baseball-diamond:0.4679762971260954,
bridge:0.39910657454410275,
ground-track-field:0.5978173014224926,
small-vehicle:0.418202632274417,
large-vehicle:0.4827424581806402,
ship:0.5763476713309732,
tennis-court:0.7866893867367266,
basketball-court:0.5900664388926606,
storage-tank:0.522142612448238,
soccer-ball-field:0.4212896657422925,
roundabout:0.5226779625553494,
harbor:0.4416480396458489,
swimming-pool:0.5307778020790656,
helicopter:0.5099261922359674,
container-crane:0.12000347981943073,
airport:0.5076951313967711,
helipad:0.1290935741372789
COCO style result:
AP50: 0.4894285127967532
AP75: 0.24352812425730352
mAP: 0.26443484834115494
The submitted information is :
Description: RetinaNet_DOTA2.0_KF_1x_20220920_52w
Username: sjtu-deter
Institute: SJTU
Emailadress: [email protected]
TeamMembers: yangxue
"""