GBDT调参
gbm算法流程图:
gbdt 参数:参考scikit-learn
The overall parameters can be divided into 3 categories:
- Tree-Specific Parameters: These affect each individual tree in the model.
- Boosting Parameters: These affect the boosting operation in the model.
- Miscellaneous Parameters: Other parameters for overall functioning.
Tree-Specific Parameters:
- min_samples_split
- Defines the minimum number of samples (or observations) which are required in a node to be considered for splitting.
- Used to control over-fitting. Higher values prevent a model from learning relations which might be highly specific to the particular sample selected for a tree.
- Too high values can lead to under-fitting hence, it should be tuned using CV.
- min_samples_leaf
- Defines the minimum samples (or observations) required in a terminal node or leaf.
- Used to control over-fitting similar to min_samples_split.
- Generally lower values should be chosen for imbalanced class problems because the regions in which the minority class will be in majority will be very small.
- min_weight_fraction_leaf
- Similar to min_samples_leaf but defined as a fraction of the total number of observations instead of an integer.
- Only one of #2 and #3 should be defined.
- max_depth
- The maximum depth of a tree.
- Used to control over-fitting as higher depth will allow model to learn relations very specific to a particular sample.
- Should be tuned using CV.
- max_leaf_nodes
- The maximum number of terminal nodes or leaves in a tree.
- Can be defined in place of max_depth. Since binary trees are created, a depth of ‘n’ would produce a maximum of 2^n leaves.
- If this is defined, GBM will ignore max_depth.
- max_features
- The number of features to consider while searching for a best split. These will be randomly selected.
- As a thumb-rule, square root of the total number of features works great but we should check upto 30-40% of the total number of features.
- Higher values can lead to over-fitting but depends on case to case.
Boosting Parameters
- learning_rate
- This determines the impact of each tree on the final outcome (step 2.4). GBM works by starting with an initial estimate which is updated using the output of each tree. The learning parameter controls the magnitude of this change in the estimates.
- Lower values are generally preferred as they make the model robust to the specific characteristics of tree and thus allowing it to generalize well.
- Lower values would require higher number of trees to model all the relations and will be computationally expensive.
- n_estimators
- The number of sequential trees to be modeled (step 2)
- Though GBM is fairly robust at higher number of trees but it can still overfit at a point. Hence, this should be tuned using CV for a particular learning rate.
- subsample
- The fraction of observations to be selected for each tree. Selection is done by random sampling.
- Values slightly less than 1 make the model robust by reducing the variance.
- Typical values ~0.8 generally work fine but can be fine-tuned further.
Miscellaneous Parameters
- loss
- It refers to the loss function to be minimized in each split.
- It can have various values for classification and regression case. Generally the default values work fine. Other values should be chosen only if you understand their impact on the model.
- init
- This affects initialization of the output.
- This can be used if we have made another model whose outcome is to be used as the initial estimates for GBM.
- random_state
- The random number seed so that same random numbers are generated every time.
- This is important for parameter tuning. If we don’t fix the random number, then we’ll have different outcomes for subsequent runs on the same parameters and it becomes difficult to compare models.
- It can potentially result in overfitting to a particular random sample selected. We can try running models for different random samples, which is computationally expensive and generally not used.
- verbose
- The type of output to be printed when the model fits. The different values can be:
- 0: no output generated (default)
- 1: output generated for trees in certain intervals
- >1: output generated for all trees
- The type of output to be printed when the model fits. The different values can be:
- warm_start
- This parameter has an interesting application and can help a lot if used judicially.
- Using this, we can fit additional trees on previous fits of a model. It can save a lot of time and you should explore this option for advanced applications
- presort
- Select whether to presort data for faster splits.
- It makes the selection automatically by default but it can be changed if needed.
#coding=utf-8
"""
http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingClassifier.html
http://scikit-learn.org/stable/modules/cross_validation.html
http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.StratifiedKFold.html#sklearn.model_selection.StratifiedKFold
http://scikit-learn.org/stable/modules/model_evaluation.html#scoring-parameter
http://scikit-learn.org/stable/modules/generated/sklearn.metrics.accuracy_score.html#sklearn.metrics.accuracy_score
"""
import pandas as pd
import numpy as np
from sklearn.ensemble import GradientBoostingClassifier
from sklearn import cross_validation, metrics
from sklearn.grid_search import GridSearchCV train = pd.read_csv('train_modified.csv')
target = 'Disbursed'
IDcol = 'ID' def modelfit(alg, dtrain, predictors, performCV = True, printFeatureImportance = True, cv_folds = 5):
alg.fit(dtrain[predictors] , dtrain['Disbursed']) # Predict class for training set x
dtrain_predictions = alg.predict(dtrain[predictors])
# predict class probabilities for training set X
dtrain_predprob = alg.predict_proba(dtrain[predictors])[:,1] #perform cross-validation
if performCV:
"""
cross validation, splitting the data, fitting a model and computing the score for cv consecutive times(with different splits each time)
cv: For integer/None inputs, if the estimator is a classifier and y is either binary or multiclass,StratifiedKFold is used.
In all other cases, KFold is used. StratifiedKFold is a variation of KFold that returns stratified folds.The folds are made
by preserving the percentage of samples for each class.
K-Folds:split dataset into k consecutive folds (without shuffling by default).Each fold is then used once as a validation
while the k - 1 remaining folds form the training set.
returns: Array of scores of the estimator for each run of the cross validation. shape = (cv,)
"""
cv_score = cross_validation.cross_val_score(alg, dtrain[predictors], dtrain['Disbursed'], cv = cv_folds, scoring = 'roc_auc') print "\nModel Report"
"""
for accuracy_score,if the parameter normalize is true, return the fraction of correctly classified samples.
If False, return the number of correctly classified samples.The best performance is 1 with normalize == True and
the number of samples with normalize == False.
"""
print "Accuracy : %.4g" % metrics.accuracy_score(dtrain['Disbursed'].values, dtrain_predictions)
"""
roc_auc_score Compute Area Under the Curve (AUC) from prediction scores
"""
print "AUC Score (Train): %f" % metrics.roc_auc_score(dtrain['Disbursed'], dtrain_predprob)
if performCV:
print "CV Score : Mean - %.7g | Std - %.7g | Min - %.7g | Max - %.7g" % (np.mean(cv_score), np.std(cv_score), np.min(cv_score), np.max(cv_score)) #print Feature importance
if printFeatureImportance:
#feature_importances_: The feature importances (the higher, the more important the feature).
feat_imp = pd.Series(alg.feature_importances_, predictors).sort_values(ascending=False) predictors = [x for x in train.columns if x not in [target, IDcol]]
gbm0 = GradientBoostingClassifier(random_state = 10)
modelfit(gbm0, train,predictors)
General Approach for Parameter Tuning
Though, GBM is robust enough to not overfit with increasing trees, but a high number for pa particular learning rate can lead to overfitting. But as we reduce the learning rate and increase trees, the computation becomes expensive and would take a long time to run on standard personal computers.
- Choose a relatively high learning rate. Generally the default value of 0.1 works but somewhere between 0.05 to 0.2 should work for different problems
- Determine the optimum number of trees for this learning rate. This should range around 40-70. Remember to choose a value on which your system can work fairly fast. This is because it will be used for testing various scenarios and determining the tree parameters.
- Tune tree-specific parameters for decided learning rate and number of trees. Note that we can choose different parameters to define a tree and I’ll take up an example here.
- Lower the learning rate and increase the estimators(增加tree的数目) proportionally to get more robust models.
Fix learning rate and number of estimators for tuning tree-based parameters
- min_samples_split = 500 : This should be ~0.5-1% of total values. Since this is imbalanced class problem, we’ll take a small value from the range.
- min_samples_leaf = 50 : Can be selected based on intuition. This is just used for preventing overfitting and again a small value because of imbalanced classes.
- max_depth = 8 : Should be chosen (5-8) based on the number of observations and predictors. This has 87K rows and 49 columns so lets take 8 here.
- max_features = ‘sqrt’ : Its a general thumb-rule to start with square root.
- subsample = 0.8 : This is a commonly used used start value
Lets take the default learning rate of 0.1 here and check the optimum number of trees for that. For this purpose, we can do a grid search and test out values from 20 to 80 in steps of 10.
predictors = [x for x in train.columns if x not in [target, IDcol]]
#gbm0 = GradientBoostingClassifier(random_state = 10)
param_test1 = {'n_estimators':range(20, 81, 10)}
#GridSearchCV exhaustive search over specified parameter values for an estimator
gridsearch1 = GridSearchCV(estimator = GradientBoostingClassifier(learning_rate = 0.1, min_samples_split = 500, min_samples_leaf = 50, max_depth = 8,
max_features = 'sqrt', subsample = 0.8, random_state = 10), param_grid = param_test1, scoring = 'roc_auc', n_jobs = 4, iid = False, cv = 5)
gridsearch1.fit(train[predictors], train[target])
print "grid_scores:", gridsearch1.grid_scores_
print "best_params:", gridsearch1.best_params_
print "best_score_:", gridsearch1.best_score_
Tuning tree-specific parameters
- Tune max_depth and num_samples_split
- Tune min_samples_leaf
- Tune max_features
The order of tuning variables should be decided carefully. You should take the variables with a higher impact on outcome first. For instance, max_depth and min_samples_split have a significant impact and we’re tuning those first.
Tuning subsample and making models with lower learning rate
注意: we need to lower the learning rate and increase the number of estimators proportionally.
参考:
https://www.zybuluo.com/yxd/note/611571
https://www.analyticsvidhya.com/blog/2016/02/complete-guide-parameter-tuning-gradient-boosting-gbm-python/
https://www.analyticsvidhya.com/blog/2016/03/complete-guide-parameter-tuning-xgboost-with-codes-python/
GBDT调参的更多相关文章
- scikit-learn 梯度提升树(GBDT)调参小结
在梯度提升树(GBDT)原理小结中,我们对GBDT的原理做了总结,本文我们就从scikit-learn里GBDT的类库使用方法作一个总结,主要会关注调参中的一些要点. 1. scikit-learn ...
- GBDT调参总结
一.GBDT类库弱学习器参数 二.回归 数据集:已知用户的30个特征,预测用户的信用值 from sklearn.ensemble import GradientBoostingRegressor f ...
- scikit-learn 梯度提升树(GBDT)调参笔记
在梯度提升树(GBDT)原理小结中,我们对GBDT的原理做了总结,本文我们就从scikit-learn里GBDT的类库使用方法作一个总结,主要会关注调参中的一些要点. 1. scikit-learn ...
- gbdt调参的小结
关键部分转自http://www.cnblogs.com/pinard/p/6143927.html 第一次知道网格搜索这个方法,不知道在工业中是不是用这种方式 1.首先从步长和迭代次数入手,选择一个 ...
- scikit-learn随机森林调参小结
在Bagging与随机森林算法原理小结中,我们对随机森林(Random Forest, 以下简称RF)的原理做了总结.本文就从实践的角度对RF做一个总结.重点讲述scikit-learn中RF的调参注 ...
- rf调参小结
转自http://www.cnblogs.com/pinard/p/6160412.html 1. scikit-learn随机森林类库概述 在scikit-learn中,RF的分类类是RandomF ...
- sklearn-GBDT 调参
1. scikit-learn GBDT类库概述 在sacikit-learn中,GradientBoostingClassifier为GBDT的分类类, 而GradientBoostingRegre ...
- xgboost/gbdt在调参时为什么树的深度很少就能达到很高的精度?
问题: 用xgboost/gbdt在在调参的时候把树的最大深度调成6就有很高的精度了.但是用DecisionTree/RandomForest的时候需要把树的深度调到15或更高.用RandomFore ...
- LightGBM 调参方法(具体操作)
sklearn实战-乳腺癌细胞数据挖掘(博主亲自录制视频) https://study.163.com/course/introduction.htm?courseId=1005269003& ...
随机推荐
- Python编程快速上手--实践项目11.11.1
from selenium import webdriver from selenium.webdriver.common.keys import Keys import time def messa ...
- Linux poll机制
1.用户空间调用(参考 poll(2) - Linux man page) int poll(struct pollfd *fds, nfds_t nfds, int timeout); it wai ...
- MPEG-4与H.264的区别 , 编码 以及 应用
MPEG4是适用于监控领域的压缩技术 MPEG4于1998年11月公布,原预计1999 年1月投入使用的国际标准MPEG4不仅是针对一定比特率下的视频.音频编码,更加注重多媒体系统的交互性和灵活性.M ...
- cf 1016D
D. Vasya And The Matrix time limit per test 2 seconds memory limit per test 256 megabytes input stan ...
- 分析nginx日志常用的命令总结
1. 利用grep ,wc命令统计某个请求或字符串出现的次数 比如统计GET /app/kevinContent接口在某天的调用次数,则可以使用如下命令: cat /usr/local/nginx/l ...
- 【2018.9.20】JOI 2017 Final T2「準急電車 / Semiexpress」
题目描述 JOI 铁路公司是 JOI 国唯一的铁路公司. 在某条铁路沿线共有 $N$ 座车站,依次编号为 $1...N$. 目前,正在服役的车次按照运行速度可分为两类:高速电车(简称快车)与普通电车( ...
- linux 中 stat 函数的用途和使用方法
stat 函数讲解 表头文件: #include <sys/stat.h> #include <unistd.h> 定义函数: int stat(const cha ...
- servlet分析
Servlet生命周期分为三个阶段: 1,初始化阶段 调用init()方法 2,响应客户请求阶段 调用service()方法 3,终止阶段 调用destroy()方法 Servlet初始化阶段: 在 ...
- json格式前端显示
使用angular可以稍加修改. [转]http://web.jobbole.com/82865/ function output(inp) { document.body.appendChild(d ...
- __new__ 与 __init__的区别
原文地址:http://www.cnblogs.com/ifantastic/p/3175735.html 单例模式 class Singleton(object): def __new__(cls, ...