Trying to increase the speed and efficiency of SIPSim



In [15]:

    
import os,sys
import numpy as np
import pandas as pd
import scipy.stats as stats
from collections import defaultdict









    



/opt/anaconda/lib/python2.7/site-packages/pytz/__init__.py:29: UserWarning: Module argparse was already imported from /opt/anaconda/lib/python2.7/argparse.pyc, but /opt/anaconda/lib/python2.7/site-packages is being added to sys.path
  from pkg_resources import resource_stream

Distributions



In [8]:

    
%%timeit

np.random.normal(size=1)









    



1000000 loops, best of 3: 1.62 µs per loop



In [9]:

    
%%timeit

stats.norm().rvs()









    



1000 loops, best of 3: 680 µs per loop

dict of pd.DataFrames versus 3d numpy array



In [26]:

    
ddf = {'a': pd.DataFrame({'a':range(10), 'b':range(10)}),
       'b': pd.DataFrame({'a':range(10), 'b':range(10)})
       }

narr = np.array([np.array(range(10) + range(10)).reshape(2,10).transpose(),
                np.array(range(10) + range(10)).reshape(2,10).transpose()
                ])



In [27]:

    
ddf









    Out[27]:





{'a':    a  b
 0  0  0
 1  1  1
 2  2  2
 3  3  3
 4  4  4
 5  5  5
 6  6  6
 7  7  7
 8  8  8
 9  9  9, 'b':    a  b
 0  0  0
 1  1  1
 2  2  2
 3  3  3
 4  4  4
 5  5  5
 6  6  6
 7  7  7
 8  8  8
 9  9  9}



In [28]:

    
narr









    Out[28]:





array([[[0, 0],
        [1, 1],
        [2, 2],
        [3, 3],
        [4, 4],
        [5, 5],
        [6, 6],
        [7, 7],
        [8, 8],
        [9, 9]],

       [[0, 0],
        [1, 1],
        [2, 2],
        [3, 3],
        [4, 4],
        [5, 5],
        [6, 6],
        [7, 7],
        [8, 8],
        [9, 9]]])



In [ ]: