— In the News —
Using simple deep learning tools, even photos obscured by YouTube’s blur feature (center, right) can be recognized.
Analytics will win votes this year. Here's how.
Data detective story about unmasking the top residential water user in Los Angeles. Nice tricks here.
— Sponsored Link —
Understand your users. Running a complex app? Record, analyze & watch live sessions of your users.
— Tools and Techniques —
Great introduction to stock market analysis. Covers how to obtain stock data using pandas, visualizing stock data, moving averages, developing a moving-average crossover strategy, backtesting, and benchmarking.
Part 4 of a popular series from NVIDIA about topics in deep learning. This post explores reinforcement learning. This is easy to follow and very well done.
Simulacrum is a simple way to pass in a dictionary object, with column names and corresponding data types and output a pandas DataFrame of random data. Great for creating a fake data set or testing data science scripts.
This Q/A with Hadley Wickham includes recommendations for R learning resources, useful packages, favorite blogs, and insights about where things are going.
— Resources —
Download and play with key datasets from Google Trends, curated by the News Lab at Google team.
— Data Viz —
Make visualizations using only markdown.
ggraph is an extension of ggplot2 tailored at plotting graph-like data structures (graphs, networks, trees, hierarchies...). It is not a graph plotting framework that uses ggplot2 underneath, but rather an extension of the ggplot2 API to make sense for graph data.