A library for making ridge plots of... ridges. Choose a location, get an elevation map, and tinker with it to make something beautiful. Heavily inspired from Zach Cole's beautiful art, Jake Vanderplas' examples, and Joy Division's 1979 album "Unknown Pleasures".
Uses matplotlib, SRTM.py, numpy, and scikit-image (for lake detection).
Install from github with
pip install git+git://github.com/colcarroll/ridge_map.git- I feel like I am missing something easy or obvious with lake/road/river/ocean detection, but what I've got gets me most of the way there. If you hack on the
RidgeMap.preprocessormethod and find something nice, I would love to hear about it! As far as I can tell, there is no way to change the color of a line in matplotlib dynamically. I would love to color the lines by elevation.Color by elevation!- Did you make a cool map? Open an issue with the code and I will add it to the examples.
The API allows you to download the data once, then edit the plot yourself, or allow the default processor to help you.
Plotting with all the defaults should give you a map of my favorite mountains.
from ridge_map import RidgeMap
RidgeMap().plot_map()
First you download the elevation data to get an array with shape
(num_lines, elevation_pts), then you can use the preprocessor
to automatically detect lakes, rivers, and oceans, and scale the elevations.
Finally, there are options to style the plot
rm = RidgeMap((11.098251,47.264786,11.695633,47.453630))
values = rm.get_elevation_data(num_lines=150)
values=rm.preprocess(
values=values,
lake_flatness=2,
water_ntile=10,
vertical_ratio=240)
rm.plot_map(values=values,
label='Karwendelgebirge',
label_y=0.1,
label_x=0.55,
label_size=40,
linewidth=1)
If you are plotting a town that is super into burnt orange for whatever reason, you can respect that choice.
rm = RidgeMap((-97.794285,30.232226,-97.710171,30.334509))
values = rm.get_elevation_data(num_lines=80)
rm.plot_map(values=rm.preprocess(values=values, water_ntile=12, vertical_ratio=40),
label='Austin\nTexas',
label_x=0.75,
linewidth=6,
line_color='orange')
The line color accepts a matplotlib colormap, so really feel free to go to town.
rm = RidgeMap((-123.107300,36.820279,-121.519775,38.210130))
values = rm.get_elevation_data(num_lines=150)
rm.plot_map(values=rm.preprocess(values=values, lake_flatness=3, water_ntile=50, vertical_ratio=30),
label='The Bay\nArea',
label_x=0.1,
line_color = plt.get_cmap('spring'))
You can find a good font from Google, and then get the path to the ttf file in the github repo.
If you pass a matplotlib colormap, you can specify kind="elevation" to color tops of mountains different from bottoms. ocean, gnuplot, and bone look nice.
from ridge_map import FontManager
font = FontManager('https://github.com/google/fonts/blob/master/ofl/uncialantiqua/UncialAntiqua-Regular.ttf?raw=True')
rm = RidgeMap((-156.250305,18.890695,-154.714966,20.275080), font=font.prop)
values = rm.get_elevation_data(num_lines=100)
rm.plot_map(values=rm.preprocess(values=values, lake_flatness=2, water_ntile=10, vertical_ratio=240),
label="Hawai'i",
label_y=0.85,
label_x=0.7,
label_size=60,
linewidth=2,
line_color=plt.get_cmap('ocean'),
kind='elevation')
I have been using this website. I find an area I like, draw a rectangle, then copy and paste the coordinates into the RidgeMap constructor.
rm = RidgeMap((-73.509693,41.678682,-73.342838,41.761581))
values = rm.get_elevation_data()
rm.plot_map(values=rm.preprocess(values=values, lake_flatness=2, water_ntile=2, vertical_ratio=60),
label='Kent\nConnecticut',
label_y=0.7,
label_x=0.65,
label_size=40)
You might really have to tune the water_ntile and lake_flatness to get the water right. You can set them to 0 if you do not want any water marked.
rm = RidgeMap((-71.167374,42.324286,-70.952454, 42.402672))
values = rm.get_elevation_data(num_lines=50)
rm.plot_map(values=rm.preprocess(values=values, lake_flatness=4, water_ntile=30, vertical_ratio=20),
label='Cambridge\nand Boston',
label_x=0.75,
label_size=40,
linewidth=1)
It is that pleasant kettle pond in the bottom right of this map, looking entirely comfortable with its place in Western writing and thought.
rm = RidgeMap((-71.418858,42.427511,-71.310024,42.481719))
values = rm.get_elevation_data(num_lines=100)
rm.plot_map(values=rm.preprocess(values=values, water_ntile=15, vertical_ratio=30),
label='Concord\nMassachusetts',
label_x=0.1,
label_size=30)
Of course! If you really want to put a stylized elevation map in a scientific plot you are making, I am not going to stop you, and will actually make it easier for you. Just pass an argument for ax to RidgeMap.plot_map.
import numpy as np
fig, axes = plt.subplots(ncols=2, figsize=(20, 5))
x = np.linspace(-2, 2)
y = x * x
axes[0].plot(x, y, 'o')
rm = RidgeMap()
rm.plot_map(label_size=24, background_color=(1, 1, 1), ax=axes[1])