c3ec7981fd91befe035b667bed8ea538c8853b97,pytorch/pytorchcv/models/pspnet.py,PSPNet,forward,#PSPNet#Any#,168
Before Change
x = self.backbone(x)
x = self.pool(x)
x = self.conv1(x)
x = self.dropout(x)
x = self.conv2(x)
x = F.interpolate(x, size=(h, w), mode="bilinear", align_corners=True)
return x
def get_pspnet(backbone,
After Change
x = self.pool(x)
x = self.final_block(x)
if self.aux:
y = self.aux_block(y)
return x, y
else:
return x
In pattern: SUPERPATTERN
Frequency: 3
Non-data size: 6
Instances
Project Name: osmr/imgclsmob
Commit Name: c3ec7981fd91befe035b667bed8ea538c8853b97
Time: 2019-03-20
Author: osemery@gmail.com
File Name: pytorch/pytorchcv/models/pspnet.py
Class Name: PSPNet
Method Name: forward
Project Name: NVIDIA/sentiment-discovery
Commit Name: 8030eeca74b6634f2a60168516573912aaa9cd65
Time: 2018-11-20
Author: raulp@nvidia.com
File Name: model/modeling.py
Class Name: BertForSequenceClassification
Method Name: forward
Project Name: osmr/imgclsmob
Commit Name: 8dc869c9450419461e671ac3e74a6893fa9c6572
Time: 2019-04-01
Author: osemery@gmail.com
File Name: gluon/gluoncv2/models/deeplabv3.py
Class Name: DeepLabv3
Method Name: hybrid_forward