No. |
Code |
0 |
# Libraries
|
1 |
import csv
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sn
import matplotlib.mlab as mlab
from matplotlib.pyplot import figure, show
|
2 |
# Loading Dataframe
|
3 |
dataframe = pd.read_csv('train.csv')
|
4 |
# Describe dataframe
|
5 |
dataframe.describe()
|
6 |
# Dataframe
|
7 |
dataframe
|
8 |
# Preparacion del dataframe
|
9 |
df_x = dataframe[['x']]
df_y = dataframe['y']
|
10 |
# Normalizar dataframe
|
11 |
df_nor_x = (df_x - df_x.mean())/df_x.std()
|
12 |
# x
|
13 |
df_nor_x
|
14 |
# y
|
15 |
df_nor_y = (df_y - df_y.mean())/df_y.std()
df_nor_y
|
16 |
# Dataframe de entrenamiento y de prueba
|
17 |
x_train, x_test, y_train, y_test = train_test_split(df_nor_x,df_nor_y,test_size=0.3,random_state=42)
|
18 |
# Model de regresion lineal
|
19 |
model = LinearRegression()
|
20 |
# Entrenamiento del model
|
21 |
model.fit(x_train,y_train)
|
22 |
# Prueba del modelo
|
23 |
predictions = model.predict(x_test)
|
24 |
# Evaluacion del modelo
|
25 |
# Coefficients
|
26 |
model.coef_
|
27 |
# Independent term
|
28 |
model.intercept_
|
29 |
# Mean squared error
|
30 |
mean_squared_error(y_test, predictions)
|
31 |
# Mean absolute error
|
32 |
mean_absolute_error(y_test, predictions)
|
33 |
# Variance
|
34 |
r2_score(y_test, predictions)
|
35 |
# Comparacion de los resultados
|
36 |
compare = pd.DataFrame({'Actual':y_test, 'Predicted':predictions})
|
37 |
# Valores de prueba
|
38 |
compare.Actual.head(10)
|
39 |
# Valores predichos
|
40 |
compare.Predicted.head(10)
|
41 |
# Grafica scatter
|
42 |
plt.scatter(y_test,predictions)
|
43 |
# Grafica de distribucion
|
44 |
sn.distplot(y_test - predictions)
|
45 |
# Preparacion del dataframe
|
46 |
df_x = dataframe[['x']]
df_y = dataframe['y']
|
47 |
# Normalizar dataframe
|
48 |
df_nor_x = (df_x - df_x.mean())/df_x.std()
|
49 |
# x
|
50 |
df_nor_x
|
51 |
# y
|
52 |
df_nor_y = (df_y - df_y.mean())/df_y.std()
df_nor_y
|
53 |
# Dataframe de entrenamiento y de prueba
|
54 |
x_train, x_test, y_train, y_test = train_test_split(df_nor_x,df_nor_y,test_size=0.3,random_state=42)
|
55 |
# Model de regresion lineal
|
56 |
model = LinearRegression()
|
57 |
# Entrenamiento del model
|
58 |
model.fit(x_train,y_train)
|
59 |
# Prueba del modelo
|
60 |
predictions = model.predict(x_test)
|
61 |
# Evaluacion del modelo
|
62 |
# Coefficients
|
63 |
model.coef_
|
64 |
# Independent term
|
65 |
model.intercept_
|
66 |
# Mean squared error
|
67 |
mean_squared_error(y_test, predictions)
|
68 |
# Mean absolute error
|
69 |
mean_absolute_error(y_test, predictions)
|
70 |
# Variance
|
71 |
r2_score(y_test, predictions)
|
72 |
# Comparacion de los resultados
|
73 |
compare = pd.DataFrame({'Actual':y_test, 'Predicted':predictions})
|
74 |
# Valores de prueba
|
75 |
compare.Actual.head(10)
|
76 |
# Valores predichos
|
77 |
compare.Predicted.head(10)
|
78 |
# Grafica scatter
|
79 |
plt.scatter(y_test,predictions)
|
80 |
# Grafica de distribucion
|
81 |
sn.distplot(y_test - predictions)
|
82 |
# Preparacion del dataframe
|
83 |
df_x = dataframe[['x']]
df_y = dataframe['y']
|
84 |
# Normalizar dataframe
|
85 |
df_nor_x = (df_x - df_x.mean())/df_x.std()
|
86 |
# x
|
87 |
df_nor_x
|
88 |
# y
|
89 |
df_nor_y = (df_y - df_y.mean())/df_y.std()
df_nor_y
|
90 |
# Dataframe de entrenamiento y de prueba
|
91 |
x_train, x_test, y_train, y_test = train_test_split(df_nor_x,df_nor_y,test_size=0.3,random_state=42)
|
92 |
# Model de regresion lineal
|
93 |
model = LinearRegression()
|
94 |
# Entrenamiento del model
|
95 |
model.fit(x_train,y_train)
|
96 |
# Prueba del modelo
|
97 |
predictions = model.predict(x_test)
|
98 |
# Evaluacion del modelo
|
99 |
# Coefficients
|
100 |
model.coef_
|
101 |
# Independent term
|
102 |
model.intercept_
|
103 |
# Mean squared error
|
104 |
mean_squared_error(y_test, predictions)
|
105 |
# Mean absolute error
|
106 |
mean_absolute_error(y_test, predictions)
|
107 |
# Variance
|
108 |
r2_score(y_test, predictions)
|
109 |
# Comparacion de los resultados
|
110 |
compare = pd.DataFrame({'Actual':y_test, 'Predicted':predictions})
|
111 |
# Valores de prueba
|
112 |
compare.Actual.head(10)
|
113 |
# Valores predichos
|
114 |
compare.Predicted.head(10)
|
115 |
# Grafica scatter
|
116 |
plt.scatter(y_test,predictions)
|
117 |
# Grafica de distribucion
|
118 |
sn.distplot(y_test - predictions)
|
119 |
# Preparacion del dataframe
|
120 |
df_x = dataframe[['x']]
df_y = dataframe['y']
|
121 |
# Normalizar dataframe
|
122 |
df_nor_x = (df_x - df_x.mean())/df_x.std()
|
123 |
# x
|
124 |
df_nor_x
|
125 |
# y
|
126 |
df_nor_y = (df_y - df_y.mean())/df_y.std()
df_nor_y
|
127 |
# Dataframe de entrenamiento y de prueba
|
128 |
x_train, x_test, y_train, y_test = train_test_split(df_nor_x,df_nor_y,test_size=0.3,random_state=42)
|
129 |
# Model de regresion lineal
|
130 |
model = LinearRegression()
|
131 |
# Entrenamiento del model
|
132 |
model.fit(x_train,y_train)
|
133 |
# Prueba del modelo
|
134 |
predictions = model.predict(x_test)
|
135 |
# Evaluacion del modelo
|
136 |
# Coefficients
|
137 |
model.coef_
|
138 |
# Independent term
|
139 |
model.intercept_
|
140 |
# Mean squared error
|
141 |
mean_squared_error(y_test, predictions)
|
142 |
# Mean absolute error
|
143 |
mean_absolute_error(y_test, predictions)
|
144 |
# Variance
|
145 |
r2_score(y_test, predictions)
|
146 |
# Comparacion de los resultados
|
147 |
compare = pd.DataFrame({'Actual':y_test, 'Predicted':predictions})
|
148 |
# Valores de prueba
|
149 |
compare.Actual.head(10)
|
150 |
# Valores predichos
|
151 |
compare.Predicted.head(10)
|
152 |
# Grafica scatter
|
153 |
plt.scatter(y_test,predictions)
|
154 |
# Grafica de distribucion
|
155 |
sn.distplot(y_test - predictions)
|
156 |
# Describe
|
157 |
dataframe.describe()
|
158 |
# Heatmap nulls
|
159 |
sn.heatmap(dataframe.isnull())
|