Canvas - a quick and dirty interface to matplotlib
python3 only
I do not know about you but I find matplotlib fanatstic but overwhelming. I can never remember its syntax, yet I find myself often repeating the same boilerplate code.
This simple library is not meant to be general but it allows me to produce the quick and dirty plots I often need.
canvas.py exposes a single object, Canvas, which has methods plot,hist,errorbar,ellipses,imshow, and save. These methods can be chained to overlap diffent types of plots. For example:
>>> from random import gauss
>>> from math import sin, cos
>>> from canvas import Canvas
>>> gaussian = [gauss(0,1) for i in range(1000)]
>>> Canvas('My First Image').hist(gaussian).save('img1.png')
hist data is an array of numbers. output
>>> spiral = [(x*cos(0.1*x),x*sin(0.1*x)) for x in range(0,300)]
>>> Canvas('My Second Image').plot(spiral).save('img2.png')
plot data is an array of 2-tuples, (x,y). output
>>> points = [(x,x+gauss(0,1),0.5) for x in range(20)]
>>> Canvas('My Third Image').errorbar(points).plot(points).save('img3.png')
errorbar data is an array of 3-tuples, (x,y,dy). In the example above the plot is superimposed to errorbars. output
>>> blobs = [(gauss(0,1),gauss(0,1),0.05,0.05) for i in range(100)]
>>> Canvas('My Fourth Image').ellipses(blobs).save('img4.png')
ellipses data is an array of 4-tuples, (x,y,dx,dy). output
>>> waves = [[sin(0.1*x)*cos(0.1*x*y) for x in range(20)] for y in range(20)]
>>> Canvas('My Fifth Image').imshow(waves).save('img5.png')
imshow data is a square 2D array of numbers. output
The names of the methods of the canvas objects are the same as the methods of the corresponding matplotlib axis object.
Django example
def my_image(request):
data = [gauss(0,1) for i in range(1000)]
image_data = Canvas('title').hist(data).binary()
return HttpResponse(image_data, mimetype="image/png")
web2py example
def my_image():
data = [gauss(0,1) for i in range(1000)]
response.headers['Content-type'] = 'image/png'
return Canvas('title').hist(data).binary()
Flask example
@app.route('/my_image')
def my_image():
data = [gauss(0,1) for i in range(1000)]
image_data = Canvas('title').hist(data).binary()
response.headers['Content-type'] = 'image/png'
return Response(image_data)
Arguments
Here is the full signature:
class Canvas(object):
def __init__(self,title='title',xlab='x',ylab='y',xrange=None,yrange=None): ...
def save(self,filename='plot.png'): ...
def hist(self,data,bins=20,color='blue',legend=None): ...
def plot(self,data,color='blue',style='-',width=2,legend=None): ...
def errorbar(self,data,color='black',marker='o',width=2,legend=None): ...
def ellipses(self,data,color='blue',width=0.01,height=0.01): ...
def imshow(self,data,interpolation='bilinear'): ...
Installation
You'll need numpy and matplotlib.
From source:
python setup.py install
If you want to install the dependancies using pip you need to process this way:
pip install numpy
# THEN
pip install matplotlib
Be sure to have numpy installed before installing matplotlib otherwise the installation will fail.
License
3-clause BSD license
