Developed and deployed my first Big Data Visualization Web Application prototype with #HoloViews #GeoViews #Bokeh #Datashader #Dask #Conda #Docker #PyViz #Python #Heroku - visualizing 136k+ geo data points at different zoom/pan views. Github repo: https://t.co/QpE4lXTdWo

Check out this interesting read at Data School

(Pandas) How to get topmost n records within each group - a very simple one liner solution

#Funky Vapourer #caterpillar just strolling down a #Haringey side street on Saturday. #moth cs

Recently noticed 300+ GB disk storage was being consumed by Redis Local AppData - having read a few articles, have now deleted them all - recovered tons of storage.

Super bookmark of the day: flattening a Pandas dataframe (with json structures in some columns) -

Thank you @lemon_disco and @jAtlas7 for spotting, taking and sharing photos of these two dragonflies @TCVLavenderPond yesterday. What stunning pics! A Black-tailed Skimmer and Emperor. #SouthwarkNature

Wildlife photography at Horniman Museum Garden: Spider @HornimanMuseum

Wildlife photography at Horniman Museum Garden: Bumble Bee @HornimanMuseum

Great Python / Jupyter tip:

https://t.co/xDkUjyY6kt

The Crisis of Credit Visualized - HD

https://t.co/iAq1BEj3gO

A map of #DataScience, xkcd-style! https://t.co/To1CwQoyjS

"Of the cast of characters mentioned in this story, the only ones that every business needs are decision-makers and analysts. The others you’ll only be able to use when you know exactly what you need them for."

Explainable AI won’t deliver. Here’s why. by @quaesita

Simple guide to confusion matrix terminology

Netflix Recommendations: Beyond the 5 stars (Part 2)

https://t.co/thqNpZOJ5N

https://t.co/RaQponFxKb

Decision Tree example illustrates how Gradient Boosting work in each iteration: (1) make predictions, (2) compute residuals, (3) update predictions with residuals (that reduce residuals and make predictions more accurate in next iteration) by @groverpr4