LTC details: normalized (nodes)



In [1]:

    
%matplotlib inline

import pandas as pd
import pylab
import matplotlib.pyplot as plt
import openpyxl



In [3]:

    
df = pd.read_csv("../../data/processed/result_detail_ltc.csv",delimiter=";")
df.head()



In [6]:

    
# Normalize lineal_threshold
df['ltc'] = df['ltc'].astype(float)
df['ltc'] = (df['ltc'] / 256490)
df.head()



In [7]:

    
# For each node group by id_node and select the last level
df2 = df.groupby(['id_node']).last()
df2.head()



In [8]:

    
df2.groupby(['lvl']).size()









    Out[8]:





lvl
0     42129
1     63189
2     73260
3     51118
4     11516
5     14861
6       211
7       171
8        22
9         8
10        4
12        2
dtype: int64



In [9]:

    
df2[['lvl','ltc']].groupby(['lvl']).size().plot(logy=True)

pylab.title("Nodes vs levels")
pylab.xlabel("Levels")
_ = pylab.ylabel("Nodes")



In [14]:

    
def print_full(x):
    pd.set_option('display.max_rows', len(x))
    print(x)
    pd.reset_option('display.max_rows')



In [15]:

    
print_full(df2[['lvl','ltc']].groupby(['lvl']).describe())









    



                    ltc
lvl                    
0   count  42129.000000
    mean       0.000007
    std        0.000016
    min        0.000004
    25%        0.000004
    50%        0.000004
    75%        0.000004
    max        0.000752
1   count  63189.000000
    mean       0.000125
    std        0.000264
    min        0.000008
    25%        0.000016
    50%        0.000039
    75%        0.000121
    max        0.003462
2   count  73260.000000
    mean       0.001004
    std        0.001108
    min        0.000012
    25%        0.000238
    50%        0.000597
    75%        0.001353
    max        0.009022
3   count  51118.000000
    mean       0.005272
    std        0.005090
    min        0.000016
    25%        0.001552
    50%        0.003552
    75%        0.007431
    max        0.042364
4   count  11516.000000
    mean       0.004471
    std        0.005204
    min        0.000035
    25%        0.001575
    50%        0.002657
    75%        0.004936
    max        0.050942
5   count  14861.000000
    mean       0.035239
    std        0.010734
    min        0.000125
    25%        0.034317
    50%        0.034317
    75%        0.036586
    max        0.103408
6   count    211.000000
    mean       0.043693
    std        0.022972
    min        0.001513
    25%        0.035976
    50%        0.041097
    75%        0.057039
    max        0.114106
7   count    171.000000
    mean       0.053063
    std        0.022264
    min        0.009314
    25%        0.038206
    50%        0.048064
    75%        0.062104
    max        0.206690
8   count     22.000000
    mean       0.060929
    std        0.026563
    min        0.011817
    25%        0.052497
    50%        0.062254
    75%        0.079687
    max        0.108714
9   count      8.000000
    mean       0.072538
    std        0.033621
    min        0.015061
    25%        0.067301
    50%        0.087148
    75%        0.093168
    max        0.097411
10  count      4.000000
    mean       0.070088
    std        0.029425
    min        0.027818
    25%        0.061866
    50%        0.079346
    75%        0.087568
    max        0.093840
12  count      2.000000
    mean       0.103445
    std        0.020249
    min        0.089126
    25%        0.096285
    50%        0.103445
    75%        0.110604
    max        0.117763



In [11]:

    
df2[['lvl','ltc']].corr()



In [12]:

    
df2[['lvl','ltc']].corr(method="spearman")



In [13]:

    
df2[['lvl','ltc']].corr(method="kendall")



In [ ]:

	id_node	lvl	ltc
0	128656	0	4
1	128656	1	7676
2	128656	2	9455
3	128656	3	9707
4	128656	4	9747

	id_node	lvl	ltc
0	128656	0	0.000016
1	128656	1	0.029927
2	128656	2	0.036863
3	128656	3	0.037846
4	128656	4	0.038001

	lvl	ltc
id_node
2	2	0.000101
3	2	0.000234
4	5	0.055827
5	1	0.000097
6	1	0.000327