ML 프로그래밍 ( 보충 ) Scikit-Learn

Transcription

1 ML 프로그래밍 ( 보충 ) Scikit-Learn

2 Scikit-Learn? 특징 a Python module integrating classic machine learning algorithms in the tightly-knit world of scientific Python packages (NumPy, SciPy, matplotlib). Simple and efficient tools for data mining and data analysis Accessible to everybody, and reusable in various contexts Built on NumPy, SciPy, and matplotlib Open source, commercially usable - BSD license

3 Popular ML Libraries Tensorflow scikit-learn Theano Pylearn2 Pyevolve NuPIC Pattern Caffe 기타다수 ( 출처 :

4 출처 : Python Machine learning projects on GitHub, with color corresponding to commits/contributors. Bob, Iepy, Nilearn, and NuPIC have the highest such value. (

5 ML tool 로서의 scikit-learn Classification Identifying to which category an object belongs to. 활용 : Spam detection, Image recognition. 알고리즘 : SVM, nearest neighbors,random forest,... Regression Predicting a continuous-valued attribute associated with an object. 활용 : Drug response, Stock prices. Algorithms: SVR, ridge regression, Lasso,... Clustering Automatic grouping of similar objects into sets. 활용 : Customer segmentation, Grouping experiment outcomes Algorithms: k-means, spectral clustering,mean-shift,... Dimensionality reduction Reducing the number of random variables to consider. 활용 : Visualization, Increased efficiency Algorithms: PCA, feature selection, non-negative matrix factorization. Model selection Comparing, validating and choosing parameters and models. Goal: Improved accuracy via parameter tuning Modules: grid search, cross validation,metrics. Preprocessing Feature extraction and normalization. 활용 : Transforming input data such as text for use with machine learning algorithms. Modules: preprocessing, feature extraction.

6 Scikit-Learn API Reference

7 sklearn.base: Base classes and utility functions Base classes Functions sklearn.cluster: Clustering Classes Functions sklearn.cluster.bicluster: Biclustering Classes sklearn.covariance: Covariance Estimators

8 sklearn.model_selection: Model Selection Splitter Classes Splitter Functions Hyper-parameter optimizers Model validation sklearn.datasets: Datasets Loaders Samples generator

9 sklearn.decomposition: Matrix Decomposition sklearn.dummy: Dummy estimators sklearn.ensemble: Ensemble Methods partial dependence sklearn.exceptions: Exceptions and warnings

10 sklearn.feature_extraction: Feature Extraction From images From text sklearn.feature_selection: Feature Selection sklearn.gaussian_process: Gaussian Processes sklearn.isotonic: Isotonic regression sklearn.kernel_approximation Kernel Approximation

11 sklearn.kernel_ridge Kernel Ridge Regression sklearn.discriminant_analysis: Discriminant Analysis sklearn.linear_model: Generalized Linear Models sklearn.manifold: Manifold Learning

12 sklearn.metrics: Metrics Model Selection Interface Classification metrics Regression metrics Multilabel ranking metrics Clustering metrics Biclustering metrics Pairwise metrics sklearn.mixture: Gaussian Mixture Models

13 sklearn.multiclass: Multiclass and multilabel classification Multiclass and multilabel classification strategies sklearn.multioutput: Multioutput regression and classification sklearn.naive_bayes: Naive Bayes sklearn.neighbors: Nearest Neighbors

14 sklearn.neural_network: Neural network models sklearn.calibration: Probability Calibration sklearn.cross_decomposition: Cross decomposition sklearn.pipeline: Pipeline sklearn.preprocessing: Preprocessing and Normalization sklearn.random_projection: Random projection

15 sklearn.semi_supervised sklearn.svm: Semi-Supervised Learning Support Vector Machines Estimators Low-level methods sklearn.tree: Decision Trees sklearn.utils: Utilities

16 class and function reference of scikit-learn. sklearn.base: Base classes and utility functions Base classes for all estimators. base.baseestimator Base class for all estimators base.classifiermixin Mixin class for all classifiers base.clustermixin Mixin class for all cluster estimators base.regressormixin Mixin class for all regression estimators base.transformermixin Mixin class for all transformers Functions base.clone(estimator[, safe]) Constructs a new estimator with the same parameters.