-
Notifications
You must be signed in to change notification settings - Fork 181
/
Copy pathcfgs_res50_dota2.0_csl_v45.py
81 lines (66 loc) · 2.03 KB
/
cfgs_res50_dota2.0_csl_v45.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
# -*- 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 = "0,1"
NUM_GPU = len(GPU_GROUP.strip().split(','))
SAVE_WEIGHTS_INTE = 40000 * 2
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')
# bbox head
ANGLE_RANGE = 180
# loss
CLS_WEIGHT = 1.0
REG_WEIGHT = 1.0
ANGLE_WEIGHT = 2.0
REG_LOSS_MODE = None
# CSL
LABEL_TYPE = 0 # {0: gaussian_label, 1: rectangular_label, 2: pulse_label, 3: triangle_label}
RADIUS = 1
OMEGA = 10
VERSION = 'RetinaNet_DOTA2.0_CSL_2x_20210429'
"""
gaussian label, omega=10
FLOPs: 700124450; Trainable params: 34019421
This is your evaluation result for task 1:
mAP: 0.43344110339222985
ap of each class:
plane:0.7474780683469242,
baseball-diamond:0.44284186301197104,
bridge:0.3602585746970695,
ground-track-field:0.5611689093593835,
small-vehicle:0.3430365248274545,
large-vehicle:0.32718842802965487, s
hip:0.4681204038689529,
tennis-court:0.7657646572876476,
basketball-court:0.5756647918918942,
storage-tank:0.4829006246589788,
soccer-ball-field:0.38423561115882415,
roundabout:0.4765156959674269,
harbor:0.3733914917262925,
swimming-pool:0.5137861147989039,
helicopter:0.38742437003275904,
container-crane:0.10810276679841897,
airport:0.42231789203218606,
helipad:0.061743072565393804
The submitted information is :
Description: RetinaNet_DOTA2.0_CSL_2x_20210429_104w
Username: sjtu-deter
Institute: SJTU
Emailadress: [email protected]
TeamMembers: yangxue
"""