Note
You are reading the documentation for MMSelfSup 0.x, which will soon be deprecated by the end of 2022. We recommend you upgrade to MMSelfSup 1.0.0rc versions to enjoy fruitful new features and better performance brought by OpenMMLab 2.0. Check out the changelog, code and documentation of MMSelfSup 1.0.0rc for more details.
Source code for mmselfsup.models.necks.odc_neck
# Copyright (c) OpenMMLab. All rights reserved.
import torch.nn as nn
from mmcv.cnn import build_norm_layer
from mmcv.runner import BaseModule
from ..builder import NECKS
[docs]@NECKS.register_module()
class ODCNeck(BaseModule):
"""The non-linear neck of ODC: fc-bn-relu-dropout-fc-relu.
Args:
in_channels (int): Number of input channels.
hid_channels (int): Number of hidden channels.
out_channels (int): Number of output channels.
with_avg_pool (bool): Whether to apply the global
average pooling after backbone. Defaults to True.
norm_cfg (dict): Dictionary to construct and config norm layer.
Defaults to dict(type='SyncBN').
init_cfg (dict or list[dict], optional): Initialization config dict.
"""
def __init__(self,
in_channels,
hid_channels,
out_channels,
with_avg_pool=True,
norm_cfg=dict(type='SyncBN'),
init_cfg=[
dict(
type='Constant',
val=1,
layer=['_BatchNorm', 'GroupNorm'])
]):
super(ODCNeck, self).__init__(init_cfg)
self.with_avg_pool = with_avg_pool
if with_avg_pool:
self.avgpool = nn.AdaptiveAvgPool2d((1, 1))
self.fc0 = nn.Linear(in_channels, hid_channels)
self.bn0 = build_norm_layer(
dict(**norm_cfg, momentum=0.001, affine=False), hid_channels)[1]
self.fc1 = nn.Linear(hid_channels, out_channels)
self.relu = nn.ReLU(inplace=True)
self.dropout = nn.Dropout()
[docs] def forward(self, x):
assert len(x) == 1
x = x[0]
if self.with_avg_pool:
x = self.avgpool(x)
x = x.view(x.size(0), -1)
x = self.fc0(x)
x = self.bn0(x)
x = self.relu(x)
x = self.dropout(x)
x = self.fc1(x)
x = self.relu(x)
return [x]