DeLF-pytorch icon indicating copy to clipboard operation
DeLF-pytorch copied to clipboard

What's mean of receptive field, stride and padding of FeatureExtractor class in extractor.py?

Open CTianyou opened this issue 6 years ago • 1 comments

I wanna extract delf with another network.How to set receptive field, stride and padding of FeatureExtractor class in extractor.py?

CTianyou avatar Dec 09 '19 10:12 CTianyou

Maybe you could set on feeder.py:Feeder as:

        if delf_config.arch in ['resnet18']:
            if self.target_layer in ['layer3']:
                self.fmap_depth = 256
                self.rf = 211  # 291.0
                self.stride = 16.0
                self.padding = 105.0
            elif self.target_layer in ['layer4']:
                self.fmap_depth = 512
                self.rf = 435.0
                self.stride = 32.0
                self.padding = 217.0

AbyssGaze avatar Jan 20 '20 02:01 AbyssGaze