vgg13bn_8xb32_in1k_gcr8.py

_base_ = ['../_base_/default_runtime.py']

# from '../_base_/models/vgg13bn.py',
# model settings
model = dict(
    type='ImageClassifier',
    backbone=dict(
        type='VGG_',  # disable the last dropout
        depth=13,
        norm_cfg=dict(type='BN'),
        num_classes=1000),
    neck=None,
    head=dict(
        type='GrassmannClsHead',
        gamma=25.0,
        dims=[8] * 1000,
        num_classes=1000,
        in_channels=4096,
        loss=dict(type='CrossEntropyLoss', loss_weight=1.0),
        topk=(1, 5),
    ))

# from '../_base_/datasets/imagenet_bs32_pil_resize.py',
# dataset settings
dataset_type = 'ImageNet'
data_preprocessor = dict(
    num_classes=1000,
    # RGB format normalization parameters
    mean=[123.675, 116.28, 103.53],
    std=[58.395, 57.12, 57.375],
    # convert image from BGR to RGB
    to_rgb=True,
)

train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='RandomResizedCrop', scale=224, backend='pillow'),
    dict(type='RandomFlip', prob=0.5, direction='horizontal'),
    dict(type='PackInputs'),
]

test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='ResizeEdge', scale=256, edge='short', backend='pillow'),
    dict(type='CenterCrop', crop_size=224),
    dict(type='PackInputs'),
]

train_dataloader = dict(
    batch_size=32,
    num_workers=5,
    dataset=dict(
        type=dataset_type,
        data_root='data/imagenet',
        ann_file='meta/train.txt',
        data_prefix='train',
        pipeline=train_pipeline),
    sampler=dict(type='DefaultSampler', shuffle=True),
)

val_dataloader = dict(
    batch_size=32,
    num_workers=5,
    dataset=dict(
        type=dataset_type,
        data_root='data/imagenet',
        ann_file='meta/val.txt',
        data_prefix='val',
        pipeline=test_pipeline),
    sampler=dict(type='DefaultSampler', shuffle=False),
)
val_evaluator = dict(type='Accuracy', topk=(1, 5))

# If you want standard test, please manually configure the test dataset
test_dataloader = val_dataloader
test_evaluator = val_evaluator

# from '../_base_/schedules/imagenet_bs256.py',
# optimizer
optim_wrapper = dict(
    optimizer=dict(type='RSGD', lr=0.1, momentum=0.9, weight_decay=0.0001))

# learning policy
param_scheduler = dict(
    type='MultiStepLR', by_epoch=True, milestones=[30, 60, 90], gamma=0.1)

# train, val, test setting
train_cfg = dict(by_epoch=True, max_epochs=100, val_interval=1)
val_cfg = dict()
test_cfg = dict()

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# based on the actual training batch size.
auto_scale_lr = dict(base_batch_size=256)