Library for gradient boosting tree

Question

Do you know of a good library for gradient boosting tree machine learning?

preferably:

• with good algorithms such as AdaBoost, TreeBoost, AnyBoost, LogitBoost, etc
• with configurable weak classifiers
• capable of both classification and prediction (regression)
• with all kinds of allowed signals: numbers, categories or free text
• C/C++ or Python
• opensource

So far I have found http://www.multiboost.org/home which looks good. But I wonder if there are other libraries?

Answer 1

These don't necessarily meet all your preferences, but there's also:

Treenet commercialization and extension of Jerome Friedman's original implementation. Not open-source but we've found it to work pretty well.

Refer the link to get R Packages which redirects to Gradient Descent for Regression Tasks. 

from sklearn.ensemble import AdaBoostClassifier 

from sklearn.ensemble import AdaBoostRegressor 

from sklearn.tree import DecisionTreeClassifier

import numpy as np

dt = DecisionTreeClassifier() 

clf = AdaBoostClassifier(n_estimators=100, base_estimator=dt,learning_rate=1)

clf.fit(X=np.random.rand(10, 3),

             y=np.random.randint(2, size=(10,)))

You can tune the parameters to optimize the performance of algorithms, I’ve mentioned below the key parameters for tuning:

One can also go through Gradient Boosting for more details on this.

from sklearn.ensemble import GradientBoostingClassifier #For Classification

from sklearn.ensemble import GradientBoostingRegressor #For Regression

import numpy as np

clf = GradientBoostingClassifier(n_estimators=100, learning_rate=1.0, max_depth=1)

clf.fit(X=np.random.rand(10, 3),

             y=np.random.randint(2, size=(10,)))