spline interpolation and custom sharpening kernels (scipy.ndimage.zoom and cv2.filter2D)

refresh on basic Pandas, MultiIndex, joins/merges, GroupBys, Pivot Tables, etc.

"massively parallel" har har. One day I'll have that much data. That's a lot of runs. OpenCL dependency.

refresh on NumPy, and using .argpartition() for kNN

Notes during Boulder Startup Week 2018, Data Science track and other talks.

along with the Boulder Data Science MeetUp

Going through the PyTorch documentation. First read about it here (https://arxiv.org/pdf/1502.05767.pdf)

stitch together GPX files, make a map. Have since accidentally deleted the data, doh.

using Flask, bokeh, and Heroku to make a semi-interesting Web App. Go ahead, try to break it! It won't be difficult.

new coolest unit: the semicircle.

Business data for a friend, and playing in Twitter's sandbox.

ML from JakeVdP's book

Strava allows bulk download of your activities. Python and the Google Maps API do the rest!

Fun with the Strava heatmap. /r/dataisbeautiful.

Scraping 17 years worth of runner data from the interwebs, and plotting and learning from it with Python/Pandas and scikit-learn.

parkrun is coming to Boulder! Here's a map of all of them. An unfinished project, at the moment.

interactively plotting stuff. all done from inside JupyterLab (including pushing this blog post).

I gave a "nerd talk" at the Powder Keg Brewing Company Tuesday Nerd Talks Series, about my milesplit project.

Scraping 17 years worth of runner data from the interwebs, and plotting and learning from it with Python/Pandas and scikit-learn.

Scraping 17 years worth of runner data from the interwebs, and plotting and learning from it with Python/Pandas and scikit-learn.

As part of a group project for the class 'Scientific Visualizaton', we aggregated data from the USGS dating back to 1900 to visualize precipitation patterns in Colorado.