Convolutional Neural Networks Are Beautiful

We all take our eyes for granted, we glance at an object for an instant and our brains can identify with ease. However distorted the information may be, we do a pretty good job at it.

Low light, obscured vision, poor eyesight... There are a myriad of situations where conditions are poor but still we manage to understand what an object is. Context helps, but we humans were created with sight in mind.

Computers have a harder time, but modern advances with convolutional neural networks are making this task a reality.

Computers are amazing, the neural networks and maps they create are beautiful.

Why not have an explore?

MapExtrakt makes viewing feature maps a breeze.

Catch a glimpse of how a computer can see.

MapExtrakt Usage

First import / gather your model (this does not have to be a pretrained pytorch model).

import torchvisionmodel = torchvision.models.vgg19(pretrained=True)

Import MapExtract's Feature Extractor and load in the model

from MapExtrackt import FeatureExtractorfe = FeatureExtractor(model)

Set image to be analysed - input can be PIL Image, Numpy array or filepath. We are using the path


View Layers


Example Output
View Single Cells At a Time

fe.display_from_map(layer_no=2, cell_no=4)

Example Output
Slice the class to get a range of cells (Layer 2 Cells 0-9)


Example Output
Or Export Layers To Video

fe.write_video(out_size=(1200,800), file_name="output.avi", time_for_layer=60, transition_perc_layer=0.2)


More Examples

For LOTS more - view the jupyter notebook.



It's as easy as PyPI

pip install mapextrackt

or build from source in terminal

git clone &&\
cd mapextrackt &&\
pip install -e .

Todo List

  • Add the ability to slice the class i.e FeatureExtractor[1,3]
  • Show parameters on the image
  • Fix video generation
  • Enable individual cells to be added to video
  • Add video parameters such as duration in seconds.
  • Clean up code
  • Make speed improvements


Created by me, initially to view the outputs for my own pleasure.

If anyone has any suggestions or requests please send them over I'd be more than happy to consider.

[email protected]

GitHub - lewis-morris/mapextrackt: Pytorch Feature Map Extractor
Pytorch Feature Map Extractor. Contribute to lewis-morris/mapextrackt development by creating an account on GitHub.