In [1]:

    

import pandas as pd
import numpy as np
import seaborn as sns
import matplotlib.pyplot as plt
from include.feature_lists import L0_stations, L1_stations, L2_stations, L3_stations

%matplotlib inline


    

In [2]:

    

df = pd.read_csv("item_station_date.csv", index_col=['Id'])


    

    




d:\Anaconda\envs\Deep2\lib\site-packages\numpy\lib\arraysetops.py:395: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison
  mask |= (ar1 == a)

In [ ]:

    

L0_df = df[L0_stations + ['L0_Minimum', 'L0_Maximum', 'L0_Duration']]
L1_df = df[L1_stations + ['L1_Minimum', 'L1_Maximum', 'L1_Duration']]
L2_df = df[L2_stations + ['L2_Minimum', 'L2_Maximum', 'L2_Duration']]
L3_df = df[L3_stations + ['L3_Minimum', 'L3_Maximum', 'L3_Duration']]


    

In [ ]:

    

L0_df_corr = L0_df.corr()


    

In [ ]:

    

plt.figure(figsize=(20,20))
sns.heatmap(L0_df_corr, annot=True)
plt.show()


    

In [ ]:

    

L1_df_corr = L1_df.corr()


    

In [ ]:

    

plt.figure(figsize=(20,20))
sns.heatmap(L1_df_corr, annot=True)
plt.show()


    

In [ ]:

    




    

In [ ]:

    

L2_df_corr = L2_df.corr()


    

In [ ]:

    

plt.figure(figsize=(20,20))
sns.heatmap(L2_df_corr, annot=True)
plt.show()


    

In [ ]:

    




    

In [ ]:

    

L3_df_corr = L3_df.corr()


    

In [ ]:

    

plt.figure(figsize=(20,20))
sns.heatmap(L3_df_corr, annot=True)
plt.show()


    

In [ ]:

    

L3_df.head(10)


    

In [130]:

    

time_df = df[['L0_Minimum', 'L0_Maximum', 'L0_Duration', 'L1_Minimum', 'L1_Maximum', 'L1_Duration', 'L2_Minimum', 'L2_Maximum', 'L2_Duration', 'L3_Minimum', 'L3_Maximum', 'L3_Duration', 'Minimum', 'Maximum', 'Total_Duration', 'Duration_Sum', 'Duration_Lag']]


    

In [107]:

    

time_df = df[['L0_Duration', 'L1_Duration', 'L2_Duration', 'L3_Duration', 'Duration_Sum', 'Total_Duration']]


    

In [ ]:

    

time_df.corr()


    

In [132]:

    

time_df_corr = time_df.corr()


    

In [133]:

    

plt.figure(figsize=(10,10))
sns.heatmap(time_df_corr, annot=True)
plt.show()


    

In [131]:

    

time_df.head(10)


    

    
Out[131]:







  

    

      

      
L0_Minimum
      
L0_Maximum
      
L0_Duration
      
L1_Minimum
      
L1_Maximum
      
L1_Duration
      
L2_Minimum
      
L2_Maximum
      
L2_Duration
      
L3_Minimum
      
L3_Maximum
      
L3_Duration
      
Minimum
      
Maximum
      
Total_Duration
      
Duration_Sum
      
Duration_Lag
    
    

      
Id
      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

      

    
  
  

    

      
510783
      
0.00
      
0.47
      
0.47
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.58
      
1.61
      
0.03
      
0.00
      
1.61
      
1.61
      
0.50
      
1.11
    
    

      
651542
      
0.00
      
0.23
      
0.23
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.49
      
1.53
      
0.04
      
0.00
      
1.53
      
1.53
      
0.27
      
1.26
    
    

      
108193
      
0.01
      
0.44
      
0.43
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.59
      
1.61
      
0.02
      
0.01
      
1.61
      
1.60
      
0.45
      
1.15
    
    

      
430412
      
0.01
      
0.23
      
0.22
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.50
      
1.53
      
0.03
      
0.01
      
1.53
      
1.52
      
0.25
      
1.27
    
    

      
443497
      
0.01
      
0.11
      
0.10
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.49
      
1.53
      
0.04
      
0.01
      
1.53
      
1.52
      
0.14
      
1.38
    
    

      
513365
      
0.01
      
0.23
      
0.22
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.49
      
1.53
      
0.04
      
0.01
      
1.53
      
1.52
      
0.26
      
1.26
    
    

      
517076
      
0.01
      
0.23
      
0.22
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.49
      
1.53
      
0.04
      
0.01
      
1.53
      
1.52
      
0.26
      
1.26
    
    

      
520216
      
0.01
      
0.23
      
0.22
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.49
      
1.52
      
0.03
      
0.01
      
1.52
      
1.51
      
0.25
      
1.26
    
    

      
521262
      
0.01
      
0.23
      
0.22
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.49
      
1.53
      
0.04
      
0.01
      
1.53
      
1.52
      
0.26
      
1.26
    
    

      
574529
      
0.01
      
0.23
      
0.22
      
NaN
      
NaN
      
0.0
      
NaN
      
NaN
      
0.0
      
1.49
      
1.52
      
0.03
      
0.01
      
1.52
      
1.51
      
0.25
      
1.26
    
  

In [33]:

    

df = df.sort_values(L0_stations + L1_stations + L2_stations + L3_stations)


    

In [34]:

    

df_flow = df[L0_stations + L1_stations + L2_stations + L3_stations]


    

In [35]:

    

df_flow.shape[0] / 4000


    

    
Out[35]:





591

In [36]:

    

df_flow[df_flow.notnull()] = 1
df_flow[df_flow.isnull()] = 0


    

    




d:\Anaconda\envs\Deep2\lib\site-packages\ipykernel_launcher.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy
  """Entry point for launching an IPython kernel.
d:\Anaconda\envs\Deep2\lib\site-packages\ipykernel_launcher.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy
  

In [14]:

    

img = [df_flow.loc[idx] if (idx % 4000 == 0) else for idx in df_flow.index]


    

    




  File "<ipython-input-14-082fa3c1ed29>", line 1
    img = [df_flow.loc[idx] if (idx % 4000 == 0) for idx in df_flow.idx]
                                                   ^
SyntaxError: invalid syntax

In [37]:

    

df_flow = df_flow.reset_index()
del df_flow['Id']


    

In [38]:

    

img = df_flow[df_flow.index % 4000 == 0]


    

In [39]:

    

plt.figure(figsize=(20,20))
plt.imshow(img.values)


    

    
Out[39]:





<matplotlib.image.AxesImage at 0x4dbcfef0>






    

In [ ]: