Sklearn for Machine Learning

Abstract:

Main page for recording learning sklearn from sklearn web.
It include many machine learning algorithm/application, both supervised or unsupervised learning, such as classification, regression, clustering, etc. Enjoy it!

Content:

We can roughly separate those traditional machine learning algorithms(excluding deep learning) into supervised or unsupervised learning algorithm based on if input data with/without labels.

Supervised learning

Supervised learning algorithm refer to learning algorithm request input data with label, by which the input is well label for the attribute we’re intrested in. And we usually want to know the same attribute/label for other external data without such label by learning on labeled data.

Classification

Regression

Unsupervised learning

Unsupoervised learning algorithm, on the other hand, refer to learning algorithm requested input data without label. The input data is unlabeled. We can’t neither are intrested in the such label. Instead, we intend to find some rules or infomation in these data, try to cluster or transfer them.

Clustering

  • kmeans is a unsupervised machine learning algorithm. It intend to separate unlabeled data to several clusters based on their feature. It’s intention is kind of like classification.

History:

  • 2017-08-13: create post for demonstration of k-means assumption
  • 2018-04-05: reorgnised page without transfer notebook to markdown, and post-create main page for skilearn package.