python 调整超参数算法gridsearch

Posted 2021-05-09

from sklearn.linear_model import ElasticNet
from sklearn.model_selection import GridSearchCV

## customize model
def customize_model(X:'array',y:'array',model:'scikit model',dparams:dict,njobs:int=2,ncv:int=3)->dict:
    grid = GridSearchCV(estimator=model, param_grid=dparams, n_jobs=njobs, cv=ncv)
    grid.fit(X, Y)
    print('Best score = %s'%grid.best_score_)
    print('Best estimator:\n%s'%grid.best_estimator_)
    print('Best params:')
    for k, v in grid.best_params_.items():
        print('- %s: %s'%(k,v))
    return grid.best_params_

# best parameters for ElasticNet model
alphas = numpy.array([1,0.1,0.01,0.001,0.0001,0])
l1_ratio = numpy.array([1,0.1,0.01,0.001,0.0001,0])
dparams = dict(alpha=alphas,l1_ratio=l1_ratio)
model = ElasticNet(random_state=0,fit_intercept=True)
customize_model(X,y,model,dparams)

from sklearn.model_selection import ParameterGrid

grid = ParameterGrid({'criterion':['mse','friedman_mse','mae'],'max_depth':[3,4,5,6,7,8,9,10]})
for params in grid:
    print(params)
    ### simulation with each set of params
    
"""
[out]:


{'criterion': 'mse', 'max_depth': 3}
{'criterion': 'mse', 'max_depth': 4}
{'criterion': 'mse', 'max_depth': 5}
{'criterion': 'mse', 'max_depth': 6}
{'criterion': 'mse', 'max_depth': 7}
{'criterion': 'mse', 'max_depth': 8}
{'criterion': 'mse', 'max_depth': 9}
{'criterion': 'mse', 'max_depth': 10}
{'criterion': 'friedman_mse', 'max_depth': 3}
{'criterion': 'friedman_mse', 'max_depth': 4}
{'criterion': 'friedman_mse', 'max_depth': 5}
{'criterion': 'friedman_mse', 'max_depth': 6}
{'criterion': 'friedman_mse', 'max_depth': 7}
{'criterion': 'friedman_mse', 'max_depth': 8}
{'criterion': 'friedman_mse', 'max_depth': 9}
{'criterion': 'friedman_mse', 'max_depth': 10}
{'criterion': 'mae', 'max_depth': 3}
{'criterion': 'mae', 'max_depth': 4}
{'criterion': 'mae', 'max_depth': 5}
{'criterion': 'mae', 'max_depth': 6}
{'criterion': 'mae', 'max_depth': 7}
{'criterion': 'mae', 'max_depth': 8}
{'criterion': 'mae', 'max_depth': 9}
{'criterion': 'mae', 'max_depth': 10}
"""

# Grid Search for Algorithm Tuning
from pandas import read_csv
import numpy
from sklearn.linear_model import Ridge
from sklearn.model_selection import GridSearchCV
url = "https://goo.gl/vhm1eU"
names = ['preg', 'plas', 'pres', 'skin', 'test', 'mass', 'pedi', 'age', 'class']
dataframe = read_csv(url, names=names)
array = dataframe.values
X = array[:,0:8]
Y = array[:,8]
alphas = numpy.array([1,0.1,0.01,0.001,0.0001,0])
param_grid = dict(alpha=alphas)
model = Ridge()
grid = GridSearchCV(estimator=model, param_grid=param_grid)
grid.fit(X, Y)
print(grid.best_score_)
print(grid.best_params_.alpha)
print(grid.best_estimator_)