Skip to content

blaylockbk/pandas-rose

Repository files navigation

Pandas Rose

This python package adds a custom Pandas accessor to generate polar wind rose plots from a Pandas dataframe.

I don't mean to compete with the wonderful windrose package already available, but that package has a little too much complexity for what I wanted. This package is meant to provide a minimal, simple interface to making wind rose plots. This is done by using Pandas methods pd.cut and df.groupby and using Matplotlib regular polar axes.

Install

Install with pip. The requirements are only pandas, numpy, and matplotlib.

pip install pandas-rose

Usage

Pandas-rose is simple.

import pandas as pd
import rose

# df is a pandas dataframe with columns
# "wind_speed" and "wind_direction"
df = pd.DataFrame({
    "wind_speed":[1,2,3,4],
    "wind_direction":[20, 10, 190,300]
})

# Display a polar wind plot of the data
df.rose.plot()

Alt text

You can specify the pandas column to use for wind direction and wind speed. You may also change the number of sectors to bin the wind direction .

df.rose.plot(
    var_column="A",    # name of variable column
    dir_column="B",    # name of direction column
    sectors=8,         # number of sectors (direction bins)
    bins=range(0,30,5) # specify variable bins
    normed=False       # If True, values as percentage instead of counts
    colors='Blues'     # Name of matplotlib colormap or list of colors
    )

There are two other accessors that give some information.

# Display a dataframe of the binned values
df.rose.table(sectors=8)

Alt text

# Display the binned data as bar graph on regular axes.
df.rose.bar()

Alt text