In [40]:

    

%matplotlib inline
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import sklearn
import sklearn.model_selection
import sklearn.linear_model
import sklearn.preprocessing
import sklearn.metrics


    

In [41]:

    

flying_animal_data_filename = 'data/report-201707170745.xls'
SPECIES_LBL = 'Species'
FLIES_LBL = 'flies'
MASS_LBL = 'Mass - average - g'
SQRT_MASS_LBL = 'sqrt mass'
LENGTH_LBL = 'Length - average - mm'
WINGSPAN_LBL = 'Wingspan - average - mm'


    

In [42]:

    

with open(flying_animal_data_filename, 'r') as infile:
    df = pd.read_excel(infile)


    

In [43]:

    

print "Dataset Rows", len(df)
flying_animals = df[df[FLIES_LBL] == 'YES'].dropna()
flying_animals[SQRT_MASS_LBL] = np.sqrt(flying_animals[MASS_LBL])
print "Flying animal Rows", len(flying_animals)
train_flying_animals, test_flying_animals = sklearn.model_selection.train_test_split(flying_animals, test_size=0.2, random_state=34)


    

    




Dataset Rows 437
Flying animal Rows 407

In [44]:

    

# Look at extreme
train_flying_animals.sort(columns=MASS_LBL, ascending=False).head(n=2)


    

    




/anaconda/lib/python2.7/site-packages/ipykernel/__main__.py:2: FutureWarning: sort(columns=....) is deprecated, use sort_values(by=.....)
  from ipykernel import kernelapp as app






    
Out[44]:






  

    

      

      
Species
      
flies
      
Mass - average - g
      
Length - average - mm
      
Wingspan - average - mm
      
sqrt mass
    
  
  

    

      
71
      
Ardeotis kori
      
YES
      
12250.0
      
1165.0
      
680.0
      
110.679718
    
    

      
148
      
Cygnus buccinator
      
YES
      
11500.0
      
1500.0
      
2200.0
      
107.238053
    
  

In [45]:

    

train_flying_animals.plot(kind='scatter', x=MASS_LBL, y=WINGSPAN_LBL, figsize=(15,8))


    

    
Out[45]:





<matplotlib.axes._subplots.AxesSubplot at 0x11bfa38d0>






    

In [46]:

    

train_flying_animals.plot(kind='scatter', x=SQRT_MASS_LBL, y=WINGSPAN_LBL, figsize=(15,8))


    

    
Out[46]:





<matplotlib.axes._subplots.AxesSubplot at 0x11bdfd8d0>






    

In [47]:

    

train_flying_animals.plot(kind='scatter', x=LENGTH_LBL, y=WINGSPAN_LBL, figsize=(15,8))


    

    
Out[47]:





<matplotlib.axes._subplots.AxesSubplot at 0x11c533d50>






    

In [48]:

    

# Model with two features
feature_names = [SQRT_MASS_LBL, LENGTH_LBL]
output_name = WINGSPAN_LBL
model = sklearn.linear_model.LinearRegression()
model.fit(train_flying_animals[feature_names], train_flying_animals[output_name])
print "Train Score", model.score(train_flying_animals[feature_names], train_flying_animals[output_name])
print "Test Score", model.score(test_flying_animals[feature_names], test_flying_animals[output_name])


    

    




Train Score 0.764003424108
Test Score 0.736355894894

In [54]:

    

print model.coef_


    

    




[ 14.23247084   0.71236716]

In [53]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [ ]:

    




    

In [ ]: