In [1]:

    

import os
import glob
import LatLon 
import numpy as np
import pandas as pd
pd.set_option('display.max_rows', 5)

# plot
%matplotlib inline
import pylab
import seaborn as sns
sns.set_style("whitegrid")
from pysurvey.plot import setup, legend, icolorbar, density

# date
from dateutil import parser
from matplotlib.dates import date2num

# database
import dataset
from sqlalchemy import create_engine
from sqlalchemy_utils import database_exists, create_database

import sompy as SOM


    

In [2]:

    

before = pd.DataFrame.from_csv('/Users/ajmendez/tmp/flight/flight_before.csv')
after = pd.DataFrame.from_csv('/Users/ajmendez/tmp/flight/flight_after.csv')


    

In [3]:

    

clean_tags = ['flight', 'date', 'time', 'weekday', 'normtime', #'datenum', 
#               'alt', 'distance', 'heading_deg', 'heading', 'x', 'y'
             ]
input_tags = ['lat', 'lon', 'alt', 'hour', 'flightnum']
before_simple = before[before['alt'] < 20000].drop(clean_tags, axis=1)
after_simple = after[after['alt'] < 20000].drop(clean_tags, axis=1)

before_simple


    

    
Out[3]:






  

    

      

      
alt
      
lat
      
lon
      
flightnum
      
datenum
      
hour
      
heading_deg
      
distance
      
heading
      
x
      
y
    
  
  

    

      
555452
      
19975
      
39.90788
      
-76.57513
      
2804
      
4.083520
      
2.004485
      
3.696297
      
64.091509
      
0.064513
      
-42.303614
      
48.146918
    
    

      
555453
      
19975
      
39.90756
      
-76.57704
      
2804
      
4.083526
      
2.004627
      
3.552561
      
64.045640
      
0.062004
      
-42.393904
      
48.006259
    
    

      
...
      
...
      
...
      
...
      
...
      
...
      
...
      
...
      
...
      
...
      
...
      
...
    
    

      
870159
      
16175
      
39.10835
      
-77.00293
      
2065
      
5.926931
      
22.246340
      
232.990672
      
41.103616
      
-2.216731
      
-5.713891
      
-40.704529
    
    

      
870160
      
16225
      
39.10779
      
-77.00024
      
2065
      
5.926944
      
22.246651
      
232.724426
      
40.956478
      
-2.221378
      
-5.504905
      
-40.584839
    
  

85293 rows × 11 columns

In [4]:

    

somdata = np.zeros( (len(before_simple), len(input_tags)) )
for i, tag in enumerate(input_tags):
    somdata[:,i] = before_simple[tag]


    

In [5]:

    

nz = 100
sm = SOM.SOM('sm', somdata, names=input_tags, 
             mapsize=(nz, nz), norm_method = 'var', initmethod='pca', neigh='Guassian')


    

In [6]:

    

sm.train(n_job=1, shared_memory='no',verbose='on')


    

    




initialization method = pca, initializing..

initialization done in 0.195000 seconds

rough training...
radius_ini: 13.000000 , radius_final: 3.250000, trainlen: 4

epoch: 1 ---> elapsed time:  9.294000, quantization error: 1.543030 

epoch: 2 ---> elapsed time:  8.939000, quantization error: 1.291100 

epoch: 3 ---> elapsed time:  8.893000, quantization error: 0.880868 

epoch: 4 ---> elapsed time:  9.130000, quantization error: 0.655456 


finetune training...
radius_ini: 3.250000 , radius_final: 1.000000, trainlen: 5

epoch: 1 ---> elapsed time:  9.301000, quantization error: 0.448378 

epoch: 2 ---> elapsed time:  8.616000, quantization error: 0.400222 

epoch: 3 ---> elapsed time:  9.217000, quantization error: 0.340749 

epoch: 4 ---> elapsed time:  9.512000, quantization error: 0.278955 

epoch: 5 ---> elapsed time:  9.679000, quantization error: 0.212927 

Total time elapsed: 104.610000 secodns
final quantization error: 0.212927

Total time elapsed: 104.611000 secodns
final quantization error: 0.212927

In [7]:

    

sm.view_map(text_size=12, COL_SiZe=3)


    

In [14]:

    

index = sm.ind_to_xy(sm.project_data(somdata))


    

In [ ]: