How to use the pylab API vs. the matplotlib API

This article has a good description of the 2 API's in matplotlib: the pylab API and the matplotlib API. I've been using the pylab interface because it is easier, especially coming from a matlab background. But I wanted to get direct access to the matplotlib classes so I needed to use the matplotlib API. Here is a simple example that creates a .png figure using the 2 different API's.

Here is the example using the pylab API:

""" api_pylab.py          
 """                       
 from pylab import *       
 
 figure(figsize=[4,4])     
 axes([.1,.1,.8,.8])       
 scatter([1,2],[3,4])      
 savefig('api_pylab.png')  
 

Here is the example using the matplotlib API:

""" api_matplotlib.py                                        
 """                                                          
 from matplotlib.figure import Figure                         
 from matplotlib.backends.backend_agg import FigureCanvasAgg  
 
 fig = Figure(figsize=[4,4])                                  
 ax = fig.add_axes([.1,.1,.8,.8])                             
 ax.scatter([1,2], [3,4])                                     
 canvas = FigureCanvasAgg(fig)                                
 canvas.print_figure("api_matplotlib.png")