About the mAP

[facias914]

This is a great piece of work. But when I copied the code, I didn't achieve the accuracy in the paper.
In fact, I ran 40 epochs on the DOTA-v1.5 data set according to the parameters in the paper, and the final accuracy achieved was only 64.9.
Can you help me see if there are any relevant parameters that need to be changed?

angle_version = 'le135'
model = dict(
type='RotatedRetinaNet',
backbone=dict(
type='ResNet',
depth=50,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
zero_init_residual=False,
norm_cfg=dict(type='BN', requires_grad=True),
norm_eval=True,
style='pytorch',
init_cfg=dict(type='Pretrained', checkpoint='torchvision://resnet50')),
neck=dict(
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
start_level=1,
add_extra_convs='on_input',
num_outs=5),
bbox_head=dict(
type='RDCFLHead',
num_classes=16,
in_channels=256,
stacked_convs=4,
feat_channels=256,
assign_by_circumhbbox=None,
dcn_assign = True,
dilation_rate = 3,
anchor_generator=dict(
type='RotatedAnchorGenerator',
octave_base_scale=4,
scales_per_octave=1,
ratios=[1.0],
strides=[8, 16, 32, 64, 128]),
bbox_coder=dict(
type='DeltaXYWHAOBBoxCoder',
angle_range=angle_version,
norm_factor=1,
edge_swap=False,
proj_xy=True,
target_means=(.0, .0, .0, .0, .0),
target_stds=(1.0, 1.0, 1.0, 1.0, 1.0)),
loss_cls=dict(
type='FocalLoss',
use_sigmoid=True,
gamma=2.0,
alpha=0.25,
loss_weight=1.0),
reg_decoded_bbox=True,
loss_bbox=dict(
type='RotatedIoULoss',
loss_weight=1.0)),
train_cfg=dict(
assigner=dict(
type='C2FAssigner',
ignore_iof_thr=-1,
gpu_assign_thr= 1024,
iou_calculator=dict(type='RBboxMetrics2D'),
assign_metric='gjsd',
topk=16,
topq=12,
constraint='dgmm',
gauss_thr=0.6),
allowed_border=-1,
pos_weight=-1,
debug=False),
test_cfg=dict(
nms_pre=2000,
min_bbox_size=0,
score_thr=0.05,
nms=dict(iou_thr=0.4),
max_per_img=2000))

dataset_type = 'DOTAv1_5Dataset'
data_root = '/data/facias/DOTA1/'
data_root1 = '/data/facias/DOTA/'
img_norm_cfg = dict(
mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='LoadAnnotations', with_bbox=True),
dict(type='RResize', img_scale=(1024, 1024)),
dict(
type='RRandomFlip',
flip_ratio=[0.25, 0.25, 0.25],
direction=['horizontal', 'vertical', 'diagonal'],
version=angle_version),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels'])
]
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(1024, 1024),
flip=False,
transforms=[
dict(type='RResize'),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img'])
])
]
data = dict(
samples_per_gpu=2,
workers_per_gpu=2,
train=dict(
type=dataset_type,
version=angle_version,
ann_file=data_root + 'trainval_split/labelTxt/',
img_prefix=data_root + 'trainval_split/images/',
pipeline=train_pipeline),
val=dict(
type=dataset_type,
version=angle_version,
ann_file=data_root + 'trainval_split/labelTxt/',
img_prefix=data_root + 'trainval_split/images/',
pipeline=test_pipeline),
test=dict(
type=dataset_type,
version=angle_version,
ann_file=data_root1 + 'test_split/images/',
img_prefix=data_root1 + 'test_split/images/',
pipeline=test_pipeline))

evaluation = dict(interval=40, metric='mAP')

optimizer = dict(type='SGD', lr=0.0025, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))

lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[24, 32, 38])
runner = dict(type='EpochBasedRunner', max_epochs=40)
checkpoint_config = dict(interval=4)

log_config = dict(
interval=50,
hooks=[
dict(type='TextLoggerHook'),
# dict(type='TensorboardLoggerHook')
])

dist_params = dict(backend='nccl')
log_level = 'INFO'
load_from = None
resume_from = None
workflow = [('train', 1)]

opencv_num_threads = 0
mp_start_method = 'fork'