mmpose/projects/skps/configs/td-hm_hrnetv2-w18_skps-1xb1...

177 lines
4.9 KiB
Python

custom_imports = dict(imports=['custom_codecs', 'models'])
_base_ = ['mmpose::_base_/default_runtime.py']
# runtime
train_cfg = dict(max_epochs=160, val_interval=1)
# optimizer
optim_wrapper = dict(
optimizer=dict(type='AdamW', lr=2e-3, weight_decay=0.0005))
# learning policy
param_scheduler = [
dict(
type='LinearLR', begin=0, end=500, start_factor=0.001,
by_epoch=False), # warm-up
dict(
type='MultiStepLR',
begin=0,
end=160,
milestones=[80, 120],
gamma=0.1,
by_epoch=True)
]
# automatically scaling LR based on the actual training batch size
auto_scale_lr = dict(base_batch_size=512)
# hooks
default_hooks = dict(checkpoint=dict(save_best='NME', rule='less', interval=1))
# codec settings
codec = dict(
type='SKPSHeatmap', input_size=(256, 256), heatmap_size=(64, 64), sigma=2)
# model settings
model = dict(
type='TopdownPoseEstimator',
data_preprocessor=dict(
type='PoseDataPreprocessor',
mean=[123.675, 116.28, 103.53],
std=[58.395, 57.12, 57.375],
bgr_to_rgb=True),
backbone=dict(
type='HRNet',
in_channels=3,
extra=dict(
stage1=dict(
num_modules=1,
num_branches=1,
block='BOTTLENECK',
num_blocks=(4, ),
num_channels=(64, )),
stage2=dict(
num_modules=1,
num_branches=2,
block='BASIC',
num_blocks=(4, 4),
num_channels=(18, 36)),
stage3=dict(
num_modules=4,
num_branches=3,
block='BASIC',
num_blocks=(4, 4, 4),
num_channels=(18, 36, 72)),
stage4=dict(
num_modules=3,
num_branches=4,
block='BASIC',
num_blocks=(4, 4, 4, 4),
num_channels=(18, 36, 72, 144),
multiscale_output=True),
upsample=dict(mode='bilinear', align_corners=False)),
init_cfg=dict(
type='Pretrained', checkpoint='open-mmlab://msra/hrnetv2_w18'),
),
neck=dict(
type='FeatureMapProcessor',
concat=True,
),
head=dict(
type='SKPSHead',
in_channels=270,
out_channels=29,
conv_out_channels=(270, ),
conv_kernel_sizes=(1, ),
heatmap_loss=dict(type='AdaptiveWingLoss', use_target_weight=True),
offside_loss=dict(type='AdaptiveWingLoss', use_target_weight=True),
decoder=codec),
test_cfg=dict(
flip_test=True,
flip_mode='heatmap',
shift_heatmap=True,
))
# base dataset settings
dataset_type = 'COFWDataset'
data_mode = 'topdown'
data_root = 'data/cofw/'
# pipelines
train_pipeline = [
dict(type='LoadImage'),
dict(type='GetBBoxCenterScale', padding=1),
dict(type='RandomFlip', direction='horizontal'),
dict(
type='Albumentation',
transforms=[
dict(type='RandomBrightnessContrast', p=0.5),
dict(type='HueSaturationValue', p=0.5),
dict(type='GaussianBlur', p=0.5),
dict(type='GaussNoise', p=0.1),
dict(
type='CoarseDropout',
max_holes=8,
max_height=0.2,
max_width=0.2,
min_holes=1,
min_height=0.1,
min_width=0.1,
p=0.5),
]),
dict(
type='RandomBBoxTransform',
shift_prob=0.,
rotate_factor=45,
scale_factor=(0.75, 1.25),
scale_prob=0),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='GenerateTarget', encoder=codec),
dict(type='PackPoseInputs')
]
val_pipeline = [
dict(type='LoadImage'),
dict(type='GetBBoxCenterScale', padding=1),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='PackPoseInputs')
]
# data loaders
train_dataloader = dict(
batch_size=16,
num_workers=4,
persistent_workers=True,
sampler=dict(type='DefaultSampler', shuffle=True),
dataset=dict(
type=dataset_type,
data_root=data_root,
data_mode=data_mode,
ann_file='annotations/cofw_train.json',
data_prefix=dict(img='images/'),
pipeline=train_pipeline,
))
val_dataloader = dict(
batch_size=32,
num_workers=4,
persistent_workers=True,
drop_last=False,
sampler=dict(type='DefaultSampler', shuffle=False, round_up=False),
dataset=dict(
type=dataset_type,
data_root=data_root,
data_mode=data_mode,
ann_file='annotations/cofw_test.json',
data_prefix=dict(img='images/'),
test_mode=True,
pipeline=val_pipeline,
))
test_dataloader = val_dataloader
# evaluators
val_evaluator = dict(
type='NME',
norm_mode='keypoint_distance',
)
test_evaluator = val_evaluator