grid search 超参数寻优

http://scikit-learn.org/stable/modules/grid_search.html

1. 超参数寻优方法 gridsearchCV 和 RandomizedSearchCV

2. 参数寻优的技巧进阶

2.1. Specifying an objective metric

By default, parameter search uses the score function of the estimator to evaluate a parameter setting. These are thesklearn.metrics.accuracy_score for classification and sklearn.metrics.r2_score for regression.

2.2 Specifying multiple metrics for evaluation

Multimetric scoring can either be specified as a list of strings of predefined scores names or a dict mapping the scorer name to the scorer function and/or the predefined scorer name(s).

http://scikit-learn.org/stable/modules/model_evaluation.html#multimetric-scoring

2.3 Composite estimators and parameter spaces 。pipeline 方法

http://scikit-learn.org/stable/modules/pipeline.html#pipeline

>>> from sklearn.pipeline import Pipeline

>>> from sklearn.svm import SVC

>>> from sklearn.decomposition import PCA

>>> estimators = [('reduce_dim', PCA()), ('clf', SVC())]

>>> pipe = Pipeline(estimators)

>>> pipe  # check pipe

         Pipeline(memory=None,

         steps=[('reduce_dim', PCA(copy=True,...)),

                ('clf', SVC(C=1.0,...))])

>>> from sklearn.pipeline import make_pipeline

>>> from sklearn.naive_bayes import MultinomialNB

>>> from sklearn.preprocessing import Binarizer

>>> make_pipeline(Binarizer(), MultinomialNB())

Pipeline(memory=None,

         steps=[('binarizer', Binarizer(copy=True, threshold=0.0)),

                ('multinomialnb', MultinomialNB(alpha=1.0,

                                                class_prior=None,

                                                fit_prior=True))])

>>> pipe.set_params(clf__C=10)  # 给clf 设定参数

>>> from sklearn.model_selection import GridSearchCV

>>> param_grid = dict(reduce_dim__n_components=[2, 5, 10],

...                   clf__C=[0.1, 10, 100])

>>> grid_search = GridSearchCV(pipe, param_grid=param_grid)

#!/usr/bin/env python2
# -*- coding: utf-8 -*-
"""
Created on Tue Sep 5 10:22:07 2017

@author: xinpingbao
"""

import numpy as np
from sklearn import datasets
from sklearn.linear_model import Ridge
from sklearn.model_selection import GridSearchCV
from sklearn.metrics import make_scorer

# load the diabetes datasets
dataset = datasets.load_diabetes()

X = dataset.data
y = dataset.target

# prepare a range of alpha values to test
alphas = np.array([1,0.1,0.01,0.001,0.0001,0])
# create and fit a ridge regression model, testing each alpha
model = Ridge()
grid = GridSearchCV(estimator=model, param_grid=dict(alpha=alphas)) # defaulting: sklearn.metrics.r2_score
# grid = GridSearchCV(estimator=model, param_grid=dict(alpha=alphas), scoring = 'metrics.mean_squared_error') # defaulting: sklearn.metrics.r2_score
grid.fit(X, y)

print(grid)
# summarize the results of the grid search
print(grid.best_score_)
print(grid.best_estimator_.alpha)

############################ 自定义error score函数 ############################

model = Ridge()

alphas = np.array([1,0.1,0.01,0.001,0.0001,0])
param_grid1 = dict(alpha=alphas)

def my_mse_error(real, pred):
w_high = 1.0
w_low = 1.0
weight = w_high * (real - pred < 0.0) + w_low * (real - pred >= 0.0)
mse = (np.sum((real - pred)**2 * weight) / float(len(real)))
return mse

def my_r2_score(y_true, y_pred):
nume = sum((y_true - y_pred) ** 2)
deno= sum((y_true - np.average(y_true, axis=0)) ** 2)

r2_score = 1 - (nume/deno)
return r2_score

error_score1 = make_scorer(my_mse_error, greater_is_better=False) # error less is better.
error_score2 = make_scorer(my_r2_score, greater_is_better=True) # error less is better.
#custom_scoring = {'weighted_MSE' : salesError}
grid_search = GridSearchCV(model, param_grid = param_grid1, scoring= error_score2, n_jobs=-1) #neg_mean_absolute_error
grid_result = grid_search.fit(X,y)
# summarize results
print("Best: %f using %s" % (grid_result.best_score_, grid_result.best_params_)) # learning_rate = 0.1

巴特西

grid search 超参数寻优

最新文章

热门文章