Skip to content

Internal Imagenet normalisation for pretrained resnet models #784

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 3 commits into from
Closed

Internal Imagenet normalisation for pretrained resnet models #784

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
7259d91
Internal Imagenet normalisation for pretrained resnet models
ekagra-ranjan Mar 8, 2019
2fba34e
Update resnet.py
ekagra-ranjan Mar 8, 2019
23e0618
fixed F821
ekagra-ranjan Mar 8, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 12 additions & 1 deletion torchvision/models/resnet.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,4 @@
import torch
import torch.nn as nn
import torch.utils.model_zoo as model_zoo

Expand Down Expand Up @@ -74,6 +75,7 @@ def __init__(self, inplanes, planes, stride=1, downsample=None):
self.stride = stride

def forward(self, x):

identity = x

out = self.conv1(x)
Expand All @@ -98,8 +100,9 @@ def forward(self, x):

class ResNet(nn.Module):

def __init__(self, block, layers, num_classes=1000, zero_init_residual=False):
def __init__(self, block, layers, num_classes=1000, zero_init_residual=False, tranform_input=False):
super(ResNet, self).__init__()
self.tranform_input = tranform_input
self.inplanes = 64
self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3,
bias=False)
Expand Down Expand Up @@ -147,6 +150,14 @@ def _make_layer(self, block, planes, blocks, stride=1):
return nn.Sequential(*layers)

def forward(self, x):

# imagenet normalisation
if self.transform_input:
x_ch0 = (torch.unsqueeze(x[:, 0], 1) - 0.485) / 0.229
x_ch1 = (torch.unsqueeze(x[:, 1], 1) - 0.456) / 0.224
x_ch2 = (torch.unsqueeze(x[:, 2], 1) - 0.406) / 0.225
x = torch.cat((x_ch0, x_ch1, x_ch2), 1)

x = self.conv1(x)
x = self.bn1(x)
x = self.relu(x)
Expand Down