Skip to content

Smitha28/KNN_HumanActivityRecognition

Repository files navigation

Click on this link to access a dynamic version of my Jupyter notebook: Binder

This project has 3 parts:

  1. Familiarize yourself with the problem and data
  2. Code a KNN Classifier from scratch, evaluate performance, and compare to Scikit-Learn's implementation
  3. Interpret results and explain findings.

This will include:

  1. Answering simple questions regarding the data
  2. Manipulating multiple DataFrames
  3. Coding functions to:
    1. Calculate Euclidean distance
    2. Calculate distance between many pairs of points
    3. Implement a majority voting system

-Combine the above to create a custom KNN algorithm -Use KNeighborsClassifier in sklearn

About

Code a KNN Classifier from scratch, evaluate performance & compare to Scikit-Learn's implementation. Using data from UC Irvine Machine Learning Repositiory

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published