shagunsodhani

Tweet thread from Signal's founder (that touches on some topics in the article) https://twitter.com/moxie/status/1474067549574688768

New article out on The Gradient, "Interpretability in Machine Learning: An Overview" providing a broad overview of the sub-field of machine learning interpretability; conceptual frameworks, existing research, and future directions.

+1 for both the suggestions

Haha. He seems to have a special place for js in his heart :)

You may find this js implementation pretty useful https://github.com/karpathy/tsnejs

Hi @SomewhatLikely. I have fixed it at both the place. Apologies for the confusion.

@coderdude, yes the purpose is to find sets of events that occur close to each other, physically, using MapReduce over large amounts of data. Thanks

Hey @popey456963, the implementation is to mine spatial colocation patterns. A spatial colocation pattern is a set of features that co-occur in space. For example, two crimes, say Robbery and Assault, would form a colocation pattern if they are reported together at many places. Think of spatial colocation pattern mining as association rule mining in the spatial domain.

Thanks for asking this question, I have updated the page.

Yes it is not the ideal solution but I implemented it for the sake of completeness.

Yeah does not sound totally legit to me.

Thanks!

There is one by Google : https://github.com/bslatkin/dpxdt. A friend of mine, Shashank Mehta, has been working on another similar tool called Bozo which tries to be a step ahead by providing cross-browser compatibility. The tool is not opensource yet but would be soon. He even proposed a talk around it in Jsfoo2015. https://jsfoo.talkfunnel.com/2015/45-automating-cross-browse... Will update here with the link once the tool is open-sourced.