Using LIME To Explain Keras Models

Shirin Glander shows us how to use the LIME package to explain image recognition models built from Keras:

The segmentation of an image into superpixels are an important step in generating explanations for image models. It is both important that the segmentation is correct and follows meaningful patterns in the picture, but also that the size/number of superpixels are appropriate. If the important features in the image are chopped into too many segments the permutations will probably damage the picture beyond recognition in almost all cases leading to a poor or failing explanation model. As the size of the object of interest is varying it is impossible to set up hard rules for the number of superpixels to segment into – the larger the object is relative to the size of the image, the fewer superpixels should be generated. Using plot_superpixels it is possible to evaluate the superpixel parameters before starting the time-consuming explanation function.

Fun stuff.  I’m glad that there’s a lot of work going into explaining neural networks rather than hand-waving them off as magic.

Related Posts

Defining TF-IDF

Bruno Stecanella explains the concept behind TF-IDF: TF-IDF was invented for document search and information retrieval. It works by increasing proportionally to the number of times a word appears in a document, but is offset by the number of documents that contain the word. So, words that are common in every document, such as this, what, and if, rank […]

Read More

Defining Tidy Data

John Mount shares thoughts about the concept of tidy data: A question is: is such a data set “tidy”? The paper itself claims the above definitions are “Codd’s 3rd normal form.” So, no the above table is not “tidy” under that paper’s definition. The the winner’s date of birth is a fact about the winner […]

Read More

Categories

June 2018
MTWTFSS
« May Jul »
 123
45678910
11121314151617
18192021222324
252627282930