Picasso: A Modular Framework for Visualizing the Learning Process of Neural Network Image Classifiers

Ryan Henderson; Rasmus Rothe

doi:10.5334/jors.178

Picasso: A Modular Framework for Visualizing the Learning Process of Neural Network Image Classifiers

Journal of Open Research Software

Volume 5 (2017): Issue 1

By: Ryan Henderson and Rasmus Rothe

Open Access

|Sep 2017

Abstract

Picasso is a free open-source (Eclipse Public License) web application written in Python for rendering standard visualizations useful for analyzing convolutional neural networks. Picasso ships with occlusion maps and saliency maps, two visualizations which help reveal issues that evaluation metrics like loss and accuracy might hide: for example, learning a proxy classification task. Picasso works with the Tensorflow deep learning framework, and Keras (when the model can be loaded into the Tensorflow backend). Picasso can be used with minimal configuration by deep learning researchers and engineers alike across various neural network architectures. Adding new visualizations is simple: the user can specify their visualization code and HTML template separately from the application code.

References

Warren S M Walter P A logical calculus of the ideas immanent in nervous activity The bulletin of mathematical biophysics 1943 5 4 115 133 10.1007/BF02478259 ISSN 1522-9602
Open DOI Search in Google Scholar Back to article
Yann L Léon B Yoshua B Patrick H Gradient-based learning applied to document recognition Proceedings of the IEEE 1998 86 11 2278 2324 10.1109/5.726791
Open DOI Search in Google Scholar Back to article
Yann L Fu J H Leon B Learning methods for generic object recognition with invariance to pose and lighting Computer Vision and Pattern Recognition, CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on, 2: II–104 2004 2004 IEEE 10.1109/CVPR.2004.1315150
Open DOI Search in Google Scholar Back to article
Alex K Ilya S Geoffrey E H Imagenet classification with deep convolutional neural networks Advances in neural information processing systems 2012 1097 1105
Search in Google Scholar Back to article
Zhang G P Avoiding pitfalls in neural network research IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) 2007 37 1 3 16 10.1109/TSMCC.2006.876059
Open DOI Search in Google Scholar Back to article
Chiyuan Z Samy B Moritz H Benjamin R Oriol V Understanding deep learning requires rethinking generalization arXiv preprint arXiv:1611.03530 2016 URL: https://arxiv.org/abs/1611.03530
Search in Google Scholar Back to article
Eliezer Y Artificial intelligence as a positive and negative factor in global risk Global catastrophic risks 2008 1 303 184
Search in Google Scholar Back to article
The Tesla Team A tragic loss 2016 URL: https://www.tesla.com/blog/tragic-loss. Accessed: 2017-5-12
Search in Google Scholar Back to article
Karen S Andrew Z Very deep convolutional networks for large-scale image recognition arXiv preprint arXiv:1409.1556 2014
Search in Google Scholar Back to article
Matthew D Z Rob F Visualizing and Understanding Convolutional Networks 2014 Cham Springer International Publishing 978-3-319-10590-1 818 833 10.1007/978-3-319-10590-1
Open DOI Search in Google Scholar Back to article
Karen S Andrea V Andrew Z Deep inside convolutional networks: Visualising image classification models and saliency maps CoRR, abs/1312.6034 2013 URL: http://arxiv.org/abs/1312.6034
Search in Google Scholar Back to article
Franois C K 2015 URL: https://github.com/fchollet/keras
Search in Google Scholar Back to article
Martín A TensorFlow: Large-scale machine learning on heterogeneous systems 2015 URL: http://tensorflow.org/. Software available from: http://tensorow.org/
Search in Google Scholar Back to article
Yann L Corinna C MNIST handwritten digit database 2010 URL: http://yann.lecun.com/exdb/mnist/
Search in Google Scholar Back to article
Jason Y Jeff C Anh N Thomas F Understanding neural networks through deep visualization Deep Learning Workshop, International Conference on Machine Learning (ICML) 2015 URL: https://github.com/yosinski/deep-visualization-toolbox
Search in Google Scholar Back to article
Raghavendra Kotikalapudi and contributors keras-vis 2017 https://github.com/raghakot/keras-vis
Search in Google Scholar Back to article
Bolei Z Aditya K Agata L Aude O Antonio T Learning deep features for discriminative localization Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2016 2921 2929 URL: http://cnnlocalization.csail.mit.edu/
Search in Google Scholar Back to article
Ramprasaath R S Abhishek D Ramakrishna V Michael C Devi P Dhruv B Grad-cam: Why did you say that? visual explanations from deep networks via gradient-based localization CoRR, abs/1610.02391 2016 URL: http://arxiv.org/abs/1610.02391
Search in Google Scholar Back to article
Yi Li Haozhi Q Jifeng D Xiangyang J Yichen W Fully convolutional instance-aware semantic segmentation CoRR, abs/1611.07709 2016 URL: http://arxiv.org/abs/1611.07709
Search in Google Scholar Back to article
Kaiming H Georgia G Piotr D Ross B G Mask R-CNN CoRR, abs/1703.06870 2017 URL: http://arxiv.org/abs/1703.06870
Search in Google Scholar Back to article
U.S. Army US Army operating Renault FT tanks URL: https://en.wikipedia.org/wiki/Light_tank#/media/File:FT-17-argonne-1918.gif
Search in Google Scholar Back to article