Orange脚本调用Data Mining Library

原文（英）：http://docs.orange.biolab.si/3/data-mining-library/

Orange Data Mining Library

教程

这是一个有好的关于 Orange 脚本使用的教程, 而Orange是一个基于Python 3的数据挖掘支持库。假定你已经下载和安装了 Orange， 可以从github repository 获得的版本，并且有一个可以正常运行的Python环境，注意需要python3，python3.4将是目前推荐的版本。使用Python shell，输入下面的代码：

% python
>>> import Orange
>>> Orange.version.version
'3.2.dev0+8907f35'
>>>

如果没有错误和警告信息，Orange 和 Python 就已经正确安装，可以继续下面的教程。

• The Data

• Data Input

• Saving the Data

• Exploration of the Data Domain

• Data Instances

• Orange Data Sets and NumPy

• Meta Attributes

• Missing Values

• Data Selection and Sampling

• Classification

• Learners and Classifiers

• Probabilistic Classification

• Cross-Validation

• Handful of Classifiers

• Regression

• Handful of Regressors

• Cross Validation

参考

可用的类库和方法：

• Data model (data)

• Data Storage (storage)

• Data Table (table)

• SQL table (data.sql)

• Domain description (domain)

• Variable Descriptors (variable)

• Values (value)

• Data Instance (instance)

• Data Filters (filter)

• Data Preprocessing (preprocess)

• Impute

• Discretization

• Continuization

• Normalization

• Randomization

• Remove

• Feature selection

• Preprocessors

• Classification (classification)

• Logistic Regression

• Random Forest

• Simple Random Forest

• Softmax Regression

• k-Nearest Neighbors

• Naive Bayes

• Multilayer perceptron (feed-forward neural network)

• Support Vector Machines

• Linear Support Vector Machines

• Nu-Support Vector Machines

• One Class Support Vector Machines

• Classification Tree

• Simple Tree

• Majority Classifier

• Elliptic Envelope

• Regression (regression)

• Linear Regression

• Mean

• Random Forest

• Simple Random Forest

• Regression Tree

• Clustering (clustering)

• Hierarchical (hierarchical)

• Distance (distance)

• Evaluation (evaluation)

• Sampling procedures for testing models (testing)

• Scoring methods (scoring)

• Miscellaneous (misc)

• Symbolic constants (enum)

• Distance Matrix (distmatrix)