Aula 2 Video 1 - A regressão linear múltipla



In [80]:

    
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
import numpy as np

Download Dataset: movies_multilinear_reg.csv



In [81]:

    
movies = pd.read_csv('datasets/movies_multilinear_reg.csv')



In [82]:

    
type(movies)









    Out[82]:





pandas.core.frame.DataFrame



In [83]:

    
movies.head(10)









    Out[83]:







  
    
      
      movieId
      Titulo
      Documentary
      Sci-Fi
      Mystery
      Horror
      Romance
      Thriller
      Crime
      Fantasy
      Comedy
      Animation
      Children
      Drama
      Adventure
      Duracao
      Investimento
      Bilheteria
    
  
  
    
      0
      1
      Toy Story (1995)
      0
      0
      0
      0
      0
      0
      0
      1
      1
      1
      1
      0
      1
      103.468310
      11.048216
      5623234.602
    
    
      1
      2
      Jumanji (1995)
      0
      0
      0
      0
      0
      0
      0
      1
      0
      0
      1
      0
      1
      112.337916
      14.927678
      5714951.757
    
    
      2
      3
      Grumpier Old Men (1995)
      0
      0
      0
      0
      1
      0
      0
      0
      1
      0
      0
      0
      0
      116.245732
      27.114597
      9524339.124
    
    
      3
      4
      Waiting to Exhale (1995)
      0
      0
      0
      0
      1
      0
      0
      0
      1
      0
      0
      1
      0
      120.317732
      4.994242
      6331568.779
    
    
      4
      5
      Father of the Bride Part II (1995)
      0
      0
      0
      0
      0
      0
      0
      0
      1
      0
      0
      0
      0
      110.023572
      19.142246
      6409617.277
    
    
      5
      6
      Heat (1995)
      0
      0
      0
      0
      0
      1
      1
      0
      0
      0
      0
      0
      0
      97.324845
      9.977311
      4956557.317
    
    
      6
      7
      Sabrina (1995)
      0
      0
      0
      0
      1
      0
      0
      0
      1
      0
      0
      0
      0
      88.978893
      14.257461
      4654565.066
    
    
      7
      8
      Tom and Huck (1995)
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      1
      0
      1
      78.536011
      8.871800
      3950017.325
    
    
      8
      9
      Sudden Death (1995)
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      104.934703
      29.112800
      6850971.551
    
    
      9
      10
      GoldenEye (1995)
      0
      0
      0
      0
      0
      1
      0
      0
      0
      0
      0
      0
      1
      105.899475
      3.695241
      5157865.850



In [84]:

    
movies.shape









    Out[84]:





(9125, 18)



In [85]:

    
filmes_independente = movies[movies.columns[2:17]]



In [86]:

    
type(filmes_independente)









    Out[86]:





pandas.core.frame.DataFrame



In [87]:

    
filmes_dependente = movies[movies.columns[17:]]



In [88]:

    
type(filmes_dependente)









    Out[88]:





pandas.core.frame.DataFrame



In [89]:

    
train, test, train_bilheteria, test_bilheteria = train_test_split(filmes_independente, filmes_dependente)



In [90]:

    
train.head()









    Out[90]:







  
    
      
      Documentary
      Sci-Fi
      Mystery
      Horror
      Romance
      Thriller
      Crime
      Fantasy
      Comedy
      Animation
      Children
      Drama
      Adventure
      Duracao
      Investimento
    
  
  
    
      7505
      0
      0
      0
      0
      0
      0
      1
      0
      1
      0
      0
      0
      0
      111.247325
      41.496397
    
    
      909
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      1
      0
      129.757430
      3.751061
    
    
      8308
      0
      0
      0
      0
      0
      0
      0
      0
      1
      0
      0
      0
      0
      139.792106
      32.107787
    
    
      416
      0
      0
      0
      0
      1
      0
      0
      0
      1
      0
      0
      0
      0
      120.752073
      6.473384
    
    
      4728
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      0
      96.975762
      4.128282



In [91]:

    
train.shape[0]









    Out[91]:





6843



In [92]:

    
test.shape[0]









    Out[92]:





2282



In [93]:

    
modelo = LinearRegression()
modelo.fit(train,train_bilheteria)









    Out[93]:





LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)

1,Toy Story (1995),0,0,0,0,0,0,0,1,1,1,1,0,1,103.4683096,11.04821649,5623234.602



In [94]:

    
modelo.predict([[0,0,0,0,0,0,0,1,1,1,1,0,1,103.4683096,11.04821649]])









    Out[94]:





array([[ 5828627.96965683]])



In [95]:

    
modelo.score(train, train_bilheteria)









    Out[95]:





0.82728993267807627



In [96]:

    
modelo.coef_









    Out[96]:





array([[ 42655.97503496,   6338.33598592, -13818.76605525, -23915.64582939,
         24775.47891531,  15437.82939179,  13901.13435994,  17001.12069112,
         -5758.97665963,  14581.77792254,  12879.50700613, -10960.33103084,
        -61189.27789748,  45674.3003098 ,  98955.21832726]])



In [97]:

    
modelo.intercept_









    Out[97]:





array([ 31992.49848156])



In [98]:

    
modelo.score(test, test_bilheteria)









    Out[98]:





0.82868767262936649

Zootopia

movieId,Titulo,Documentary,Sci-Fi,Mystery,Horror,Romance,Thriller,Crime,Fantasy,Comedy,Animation,Children,Drama,Adventure,Duracao,Investimento,Bilheteria
9999999,Zootopia,0,0,0,0,0,0,0,1,1,1,1,0,1,110,27.74456356,?????



In [99]:

    
zootopia = [0,0,0,0,0,0,0,1,1,1,1,0,1,110,27.74456356]
modelo.predict([zootopia])









    Out[99]:





array([[ 7779149.0280967]])

Split dos dados com test_size=0.3



In [100]:

    
train, test, train_bilheteria, test_bilheteria = train_test_split(filmes_independente, filmes_dependente, test_size=0.3)



In [101]:

    
modelo_30 = LinearRegression()
modelo_30.fit(train,train_bilheteria)









    Out[101]:





LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)



In [102]:

    
modelo_30.score(test, test_bilheteria)









    Out[102]:





0.8324125338489522



In [103]:

    
zootopia = [0,0,0,0,0,0,0,0,1,1,1,0,1,145.5170642,3.451632127]
modelo.predict([zootopia])









    Out[103]:





array([[ 6980452.63003705]])



In [104]:

    
planeta_macaco = [0,1,0,0,0,0,0,0,0,0,0,0,0,150,5]
modelo.predict([planeta_macaco])









    Out[104]:





array([[ 7384251.97257379]])

	movieId	Titulo	Romance	Thriller	Crime	Fantasy	Comedy	Animation	Children	Drama	Adventure	Duracao	Investimento	Bilheteria
0	1	Toy Story (1995)	0	0	0	1	1	1	1	0	1	103.468310	11.048216	5623234.602
1	2	Jumanji (1995)	0	0	0	1	0	0	1	0	1	112.337916	14.927678	5714951.757
2	3	Grumpier Old Men (1995)	1	0	0	0	1	0	0	0	0	116.245732	27.114597	9524339.124
3	4	Waiting to Exhale (1995)	1	0	0	0	1	0	0	1	0	120.317732	4.994242	6331568.779
4	5	Father of the Bride Part II (1995)	0	0	0	0	1	0	0	0	0	110.023572	19.142246	6409617.277
5	6	Heat (1995)	0	1	1	0	0	0	0	0	0	97.324845	9.977311	4956557.317
6	7	Sabrina (1995)	1	0	0	0	1	0	0	0	0	88.978893	14.257461	4654565.066
7	8	Tom and Huck (1995)	0	0	0	0	0	0	1	0	1	78.536011	8.871800	3950017.325
8	9	Sudden Death (1995)	0	0	0	0	0	0	0	0	0	104.934703	29.112800	6850971.551
9	10	GoldenEye (1995)	0	1	0	0	0	0	0	0	1	105.899475	3.695241	5157865.850

	Romance	Crime	Comedy	Drama	Duracao	Investimento
7505	0	1	1	0	111.247325	41.496397
909	0	0	0	1	129.757430	3.751061
8308	0	0	1	0	139.792106	32.107787
416	1	0	1	0	120.752073	6.473384
4728	0	0	0	0	96.975762	4.128282