simclr_resnet50_8xb32-coslr-200e_in1k.py

_base_ = [
    '../_base_/datasets/imagenet_bs32_simclr.py',
    '../_base_/schedules/imagenet_lars_coslr_200e.py',
    '../_base_/default_runtime.py',
]

# model settings
model = dict(
    type='SimCLR',
    backbone=dict(
        type='ResNet',
        depth=50,
        norm_cfg=dict(type='SyncBN'),
        zero_init_residual=True),
    neck=dict(
        type='NonLinearNeck',  # SimCLR non-linear neck
        in_channels=2048,
        hid_channels=2048,
        out_channels=128,
        num_layers=2,
        with_avg_pool=True),
    head=dict(
        type='ContrastiveHead',
        loss=dict(type='CrossEntropyLoss'),
        temperature=0.1),
)

# optimizer
optim_wrapper = dict(
    type='OptimWrapper',
    optimizer=dict(type='LARS', lr=0.3, momentum=0.9, weight_decay=1e-6),
    paramwise_cfg=dict(
        custom_keys={
            'bn': dict(decay_mult=0, lars_exclude=True),
            'bias': dict(decay_mult=0, lars_exclude=True),
            # bn layer in ResNet block downsample module
            'downsample.1': dict(decay_mult=0, lars_exclude=True),
        }))

# runtime settings
default_hooks = dict(
    # only keeps the latest 3 checkpoints
    checkpoint=dict(type='CheckpointHook', interval=10, max_keep_ckpts=3))

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