projects/gesture_recognition/configs/stgcnpp_8xb16-joint-u100-16e_jester-keypoint-2d.py

_base_ = 'mmaction::_base_/default_runtime.py'

hand_layout = dict(
    num_node=17,
    inward=[(15, 13), (13, 11), (16, 14), (14, 12), (11, 5), (12, 6), (9, 7),
            (7, 5), (10, 8), (8, 6), (5, 0), (6, 0), (1, 0), (3, 1), (2, 0),
            (4, 2)],
    center=0)

model = dict(
    type='RecognizerGCN',
    backbone=dict(
        type='STGCN',
        gcn_adaptive='init',
        gcn_with_res=True,
        tcn_type='mstcn',
        graph_cfg=dict(layout=hand_layout, mode='spatial')),
    cls_head=dict(type='GCNHead', num_classes=27, in_channels=256))

dataset_type = 'PoseDataset'
ann_file = 'jester.pkl'
train_pipeline = [
    dict(type='PreNormalize2D'),
    dict(type='GenSkeFeat', dataset='coco', feats=['j']),
    dict(type='UniformSampleFrames', clip_len=100),
    dict(type='PoseDecode'),
    dict(type='FormatGCNInput', num_person=1),
    dict(type='PackActionInputs')
]
val_pipeline = [
    dict(type='PreNormalize2D'),
    dict(type='GenSkeFeat', dataset='coco', feats=['j']),
    dict(
        type='UniformSampleFrames', clip_len=100, num_clips=1, test_mode=True),
    dict(type='PoseDecode'),
    dict(type='FormatGCNInput', num_person=1),
    dict(type='PackActionInputs')
]
test_pipeline = [
    dict(type='PreNormalize2D'),
    dict(type='GenSkeFeat', dataset='coco', feats=['j']),
    dict(
        type='UniformSampleFrames', clip_len=100, num_clips=10,
        test_mode=True),
    dict(type='PoseDecode'),
    dict(type='FormatGCNInput', num_person=1),
    dict(type='PackActionInputs')
]

train_dataloader = dict(
    batch_size=4,
    num_workers=2,
    persistent_workers=True,
    sampler=dict(type='DefaultSampler', shuffle=True),
    dataset=dict(
        type='RepeatDataset',
        times=5,
        dataset=dict(
            type=dataset_type,
            ann_file=ann_file,
            pipeline=train_pipeline,
            split='train')))
val_dataloader = dict(
    batch_size=16,
    num_workers=2,
    persistent_workers=True,
    sampler=dict(type='DefaultSampler', shuffle=False),
    dataset=dict(
        type=dataset_type,
        ann_file=ann_file,
        pipeline=val_pipeline,
        split='val',
        test_mode=True))
test_dataloader = dict(
    batch_size=1,
    num_workers=2,
    persistent_workers=True,
    sampler=dict(type='DefaultSampler', shuffle=False),
    dataset=dict(
        type=dataset_type,
        ann_file=ann_file,
        pipeline=test_pipeline,
        split='val',
        test_mode=True))

val_evaluator = [dict(type='AccMetric')]
test_evaluator = val_evaluator

train_cfg = dict(
    type='EpochBasedTrainLoop', max_epochs=16, val_begin=1, val_interval=1)
val_cfg = dict(type='ValLoop')
test_cfg = dict(type='TestLoop')

param_scheduler = [
    dict(
        type='CosineAnnealingLR',
        eta_min=0,
        T_max=16,
        by_epoch=True,
        convert_to_iter_based=True)
]

optim_wrapper = dict(
    optimizer=dict(
        type='SGD', lr=0.1, momentum=0.9, weight_decay=0.0005, nesterov=True))

default_hooks = dict(checkpoint=dict(interval=1), logger=dict(interval=100))

# Default setting for scaling LR automatically
#   - `enable` means enable scaling LR automatically
#       or not by default.
#   - `base_batch_size` = (8 GPUs) x (16 samples per GPU).
auto_scale_lr = dict(enable=False, base_batch_size=128)