country_name
overall_value
85 / 100
Neymar Jr7.1%
Alisson7.1%
12 others85.7%
25 - 34
2
Neymar Jr
27
13
Alisson
26
25
Ederson
25
42
Casemiro
27
44
Fernandinho
34
45
Thiago Silva
34
56
Marquinhos
25
63
Roberto Firmino
27
68
Coutinho
27
81
Fabinho
25
14 rows, showing
per page
Page of 2
club_name
overall_value
80 / 100
R. Lewandowski6.3%
M. Neuer6.3%
14 others87.5%
23 - 33
20
R. Lewandowski
30
31
M. Neuer
33
52
Thiago
28
61
J. Kimmich
24
68
Coutinho
27
70
T. Müller
29
77
N. Süle
23
92
D. Alaba
27
114
L. Hernández
23
117
L. Goretzka
24
16 rows, showing
per page
Page of 2
attributes_df.drop(['height_cm', 'weight_kg', 'weak_foot', 'pace'], axis=1, inplace=True)
lr = LinearRegression()
lr.fit(X_train, y_train)
y_pred_lr = lr.predict(X_test)
svr_rbf=SVR(C=1.0, epsilon=0.2, kernel='rbf')
svr_rbf.fit(X_train, y_train)
y_pred_svr = svr_rbf.predict(X_test)
forest=RandomForestRegressor(n_estimators=20, max_depth=10, criterion='mse')
forest.fit(X_train,y_train)
y_pred_rfr = forest.predict(X_test)
/shared-libs/python3.7/py/lib/python3.7/site-packages/sklearn/ensemble/_forest.py:392: FutureWarning: Criterion 'mse' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='squared_error'` which is equivalent.
FutureWarning,
evaluation(y_test, y_pred_lr, 'LinearRegression')
evaluation(y_test, y_pred_svr, 'SupportVectorRegressor')
evaluation(y_test, y_pred_rfr, 'RandomForestRegressor')
The metrics for LinearRegression are:
MAE: 2.420334792365451
MSE: 9.515259474609762
R2Score: 0.7962606365214295
The metrics for SupportVectorRegressor are:
MAE: 2.022662447939191
MSE: 8.875203314184791
R2Score: 0.8099654267127521
The metrics for RandomForestRegressor are:
MAE: 0.331646241704678
MSE: 0.24941789604158332
R2Score: 0.9946594999836554
new_forest=RandomForestRegressor(n_estimators=20, max_depth=3, criterion='mse')
new_forest.fit(X_train,y_train)
new_y_pred_rfr = new_forest.predict(X_test)
/shared-libs/python3.7/py/lib/python3.7/site-packages/sklearn/ensemble/_forest.py:392: FutureWarning: Criterion 'mse' was deprecated in v1.0 and will be removed in version 1.2. Use `criterion='squared_error'` which is equivalent.
FutureWarning,
evaluation(y_test, new_y_pred_rfr, 'RandomForestRegressor')
The metrics for RandomForestRegressor are:
MAE: 1.671335296427873
MSE: 5.491478228484632
R2Score: 0.8824172602109968