Created
April 17, 2019 11:08
-
-
Save RicherMans/6c38249d582b23f1ee7ed91d7d7894ca to your computer and use it in GitHub Desktop.
Pretrained Resnet used for SED
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| class PretrainedResNet(torch.nn.Module): | |
| """Docstring for PretrainedResNet. """ | |
| def __init__(self, outputdim): | |
| """TODO: to be defined1. | |
| :outputdim: TODO | |
| """ | |
| torch.nn.Module.__init__(self) | |
| import pretorched # HUIIIIIII | |
| self._outputdim = outputdim | |
| model_name = 'resnet3d50' | |
| self.net = pretorched.__dict__[model_name]( | |
| num_classes=339, pretrained='moments') | |
| self.net.last_linear = nn.Linear(2048, outputdim) | |
| self.net.avgpool = nn.AdaptiveAvgPool3d((None, 1, 1)) | |
| def forward(self, x): | |
| x = self.net.conv1(x) | |
| x = self.net.bn1(x) | |
| x = self.net.relu(x) | |
| x = self.net.maxpool(x) | |
| x = self.net.layer1(x) | |
| x = self.net.layer2(x) | |
| x = self.net.layer3(x) | |
| x = self.net.layer4(x) | |
| x = self.net.avgpool(x) | |
| x = x.transpose(1, 2).contiguous() | |
| x = x.view(x.shape[0], x.shape[1], -1) | |
| return self.net.last_linear(x) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment