Randomly generates circles that change position and size.
This is a vizualization of hourly bike availability for Seattle's Pronto! ride-sharing service for the month of July 2015.
- The size of the circles represents the number of bikes at the station
- The color represents how full the station is where red is empty, green is 50%, and bull is 100%
Click on the graph to draw points. A line of best fit will dynamically update to the new data points.
Summary statistics from the cash for clunkers dataset by state.
Please note:
- Average gallons saved and average gallons saved per car are yearly estimates.
- I used the average miles driven per year (13,476 miles), the trade in car's mpg, and the new car's mpg to find an estimate gallons saved for each participant.
I created this class because of feedback from Dr. Alexander on my first project at Galvanize. He suggested hierarchical clustering on the columns to reduce the large feature space into hopefully interpretable combinations.
Getting started is very simple:
from columnwiseclustering import CWHC
X = df.values
names = df.columns
Uses react and d3 to plot the letter fequency of a sentance as it is typed. React and D3 integration based on this blog post by Nicolas Hery and letter frequency chart based on this example by Mike Bostock.
I created this class to quickly implement versions of PCA and develope intuition through plotting and examining the principle componenents. The syntax follows scikit learn's philosophy, with a few modifications to improve the work flow for the specific uses of this class.
Getting started is very simple:
from reducedimensions import DRPC
X = df.values
names = df.columns
license: mit |