Overview of Scikit-Learn (Machine Learning in Python)

Topic:
Vivian will go over the main categories of all the popular algorithms and introduce a few user cases of 3 mostly voted methods.

The options are:

Regression: linear_model.LinearRegression, linear_model.Ridge, linear_model.Lasso, linear_model.ElasticNet

Classification(Discriminant Analysis): lda.LDA qda.QDA

Classification(Tree based model): tree.DecisionTreeClassifier ensemble.RandomForestClassifier

Classification(the others): linear_model.LogisticRegression svm.SVC

Classification(Nearest Neighbors) :neighbors.KNeighborsClassifier neighbors.RadiusNeighborsClassifier

Classification(Naive Bayes): naive_bayes.GaussianNB naive_bayes.MultinomialNB naive_bayes.BernoulliNB

Unsupervised Learning: decomposition.PCA cluster.KMeans cluster.AgglomerativeClustering

Feature Selection: feature_selection.VarianceThreshold feature_selection.SelectKBest feature_selection.SelectPercentile

Cross-Validation: cross_validation.KFold cross_validation.StratifiedKFold cross_validation.cross_val_score cross_validation.train_test_split

Model Selection: linear_model.RidgeCV linear_model.LassoCV linear_model.ElasticNetCV grid_search.GridSearchCV

Desktop Recording:

[youtube]http://youtu.be/hNwUq0Z70Zk[/youtube]