notebook.community

Edit and run



In [1]:

    
%pylab inline
import pandas as pd









    



Populating the interactive namespace from numpy and matplotlib



In [2]:

    
%ls ./competition_data/









    



bill_of_materials.csv  comp_other.csv         test_set.csv
comp_adaptor.csv       comp_sleeve.csv        train_set.csv
comp_boss.csv          comp_straight.csv      tube.csv
comp_elbow.csv         comp_tee.csv           tube_end_form.csv
comp_float.csv         comp_threaded.csv      type_component.csv
comp_hfl.csv           components.csv         type_connection.csv
comp_nut.csv           specs.csv              type_end_form.csv



In [3]:

    
train_df = pd.read_csv('./competition_data/train_set.csv')
train_df.shape









    Out[3]:





(30213, 8)



In [4]:

    
train_df.head()









    Out[4]:






  
    
      
      tube_assembly_id
      supplier
      quote_date
      annual_usage
      min_order_quantity
      bracket_pricing
      quantity
      cost
    
  
  
    
      0
      TA-00002
      S-0066
      2013-07-07
      0
      0
      Yes
      1
      21.905933
    
    
      1
      TA-00002
      S-0066
      2013-07-07
      0
      0
      Yes
      2
      12.341214
    
    
      2
      TA-00002
      S-0066
      2013-07-07
      0
      0
      Yes
      5
      6.601826
    
    
      3
      TA-00002
      S-0066
      2013-07-07
      0
      0
      Yes
      10
      4.687770
    
    
      4
      TA-00002
      S-0066
      2013-07-07
      0
      0
      Yes
      25
      3.541561



In [5]:

    
train_df.dtypes









    Out[5]:





tube_assembly_id       object
supplier               object
quote_date             object
annual_usage            int64
min_order_quantity      int64
bracket_pricing        object
quantity                int64
cost                  float64
dtype: object



In [6]:

    
train_df.describe()









    Out[6]:






  
    
      
      annual_usage
      min_order_quantity
      quantity
      cost
    
  
  
    
      count
      30213.000000
      30213.000000
      30213.000000
      30213.000000
    
    
      mean
      120.369377
      2.084699
      38.389369
      13.433317
    
    
      std
      1590.331872
      12.742776
      70.761392
      28.663200
    
    
      min
      0.000000
      0.000000
      1.000000
      0.503553
    
    
      25%
      0.000000
      0.000000
      2.000000
      3.878190
    
    
      50%
      0.000000
      0.000000
      10.000000
      6.521146
    
    
      75%
      2.000000
      0.000000
      40.000000
      13.431781
    
    
      max
      150000.000000
      535.000000
      2500.000000
      1000.000000



In [7]:

    
test_df = pd.read_csv('./competition_data/test_set.csv')



In [8]:

    
test_df.head()









    Out[8]:






  
    
      
      id
      tube_assembly_id
      supplier
      quote_date
      annual_usage
      min_order_quantity
      bracket_pricing
      quantity
    
  
  
    
      0
      1
      TA-00001
      S-0066
      2013-06-23
      0
      0
      Yes
      1
    
    
      1
      2
      TA-00001
      S-0066
      2013-06-23
      0
      0
      Yes
      2
    
    
      2
      3
      TA-00001
      S-0066
      2013-06-23
      0
      0
      Yes
      5
    
    
      3
      4
      TA-00001
      S-0066
      2013-06-23
      0
      0
      Yes
      10
    
    
      4
      5
      TA-00001
      S-0066
      2013-06-23
      0
      0
      Yes
      25



In [9]:

    
train_df.cost.describe()









    Out[9]:





count    30213.000000
mean        13.433317
std         28.663200
min          0.503553
25%          3.878190
50%          6.521146
75%         13.431781
max       1000.000000
Name: cost, dtype: float64



In [10]:

    
plt.boxplot(train_df.cost)
plt.show()



In [11]:

    
train_df[train_df.cost > 200].shape









    Out[11]:





(125, 8)



In [12]:

    
train_df[train_df.cost > 100].shape









    Out[12]:





(399, 8)



In [13]:

    
plt.hist(train_df.cost, bins=100, range=(0, 100))
plt.show()



In [14]:

    
train_df[train_df.cost==1000]









    Out[14]:






  
    
      
      tube_assembly_id
      supplier
      quote_date
      annual_usage
      min_order_quantity
      bracket_pricing
      quantity
      cost
    
  
  
    
      10364
      TA-06161
      S-0026
      2008-06-23
      1
      1
      No
      1
      1000



In [15]:

    
train_df.quantity.plot(kind='hist', bins = 100, range=(0, 100))









    Out[15]:





<matplotlib.axes._subplots.AxesSubplot at 0x10829d510>



In [38]:

    
quantity_feature = train_df.quantity



In [51]:

    
quantity_feature_val_counts = quantity_feature.value_counts()



In [59]:

    
quantity_feature_val_counts[[1,5,10,2,25,50,100,250,20]].sum()









    Out[59]:





26815



In [60]:

    
26815 / 30213.0









    Out[60]:





0.8875318571475854



In [ ]:

	tube_assembly_id	supplier	quote_date	bracket_pricing	quantity	cost
0	TA-00002	S-0066	2013-07-07	Yes	1	21.905933
1	TA-00002	S-0066	2013-07-07	Yes	2	12.341214
2	TA-00002	S-0066	2013-07-07	Yes	5	6.601826
3	TA-00002	S-0066	2013-07-07	Yes	10	4.687770
4	TA-00002	S-0066	2013-07-07	Yes	25	3.541561

	annual_usage	min_order_quantity	quantity	cost
count	30213.000000	30213.000000	30213.000000	30213.000000
mean	120.369377	2.084699	38.389369	13.433317
std	1590.331872	12.742776	70.761392	28.663200
min	0.000000	0.000000	1.000000	0.503553
25%	0.000000	0.000000	2.000000	3.878190
50%	0.000000	0.000000	10.000000	6.521146
75%	2.000000	0.000000	40.000000	13.431781
max	150000.000000	535.000000	2500.000000	1000.000000

	id	tube_assembly_id	supplier	quote_date	bracket_pricing	quantity
0	1	TA-00001	S-0066	2013-06-23	Yes	1
1	2	TA-00001	S-0066	2013-06-23	Yes	2
2	3	TA-00001	S-0066	2013-06-23	Yes	5
3	4	TA-00001	S-0066	2013-06-23	Yes	10
4	5	TA-00001	S-0066	2013-06-23	Yes	25