blob_classify_kernelized_svm

blob_classify_kernelized_svm, a scikit-learn code which uses multiple applications of logistic regression to classify an artificial dataset of three groups of "blobs".

Logistic regression can only seek a boundary that splits data into two groups. If we have three groups, we can nonetheless use logistic regression three times, finding boundarys 0 versus 1 and 2, 1 versus 0 and 2, and 2 versus 0 and 1.

Licensing:

The computer code and data files described and made available on this web page are distributed under the MIT license

Related Data and Programs:

blob_classify_logistic_multi, a scikit-learn code which uses multiple applications of logistic regression to classify an artificial dataset of three groups of "blobs".

blob_cluster_kmeans, a scikit-learn code which uses the k-means algorithm to cluster four groups of "blobs".

Reference:

• Andreas Mueller, Sarah Guido,
  Introduction to Machine Learning with Python,
  OReilly, 2017,
  ISBN: 978-1-449-36941-5

Source Code:

• blob_classify_kernelized_svm.py, the scikit-learn script.
• blob_classify_kernelized_svm.sh, runs the script.
• blob_classify_kernelized_svm.txt, the output file.

• blob_data.png, the data.
• blob_boundary.png, the data with the SVM boundary line.
• blob_data_kernelized.png, the kernelized data.
• blob_boundary_kernelized.png, the 3D data with the kernelized SVM boundary line.
• blob_boundary_projected.png, the 2D data with the projected kernelized SVM boundary line.