Skip to content

CS3244 Group 26 project - Human Activity and Postural Transition Recognition

Notifications You must be signed in to change notification settings

zunedz/CS3244-Project

Repository files navigation

CS3244 Project - HUMAN ACTIVITY AND POSTURAL TRANSITIONS RECOGNITION

Motivation

Utilize this human activity dataset to detect individuals’ actions real time and estimate the calories burnt so as to make necessary changes.

Dataset description

dataset consists of accelerometer and gyroscope 3-axial raw signals readings from user smartphones that are placed in their waist while performing these 12 activities (6 basic activities: standing, sitting lying, walking, walking downstairs and walking upstairs and 6 postural transitions: stand-to-sit, sit-to-stand, sit-to-lie, lie-to-sit, stand-to-lie, lie-to-stand). The dataset has been pre-processed and converted from time series data into tabular data with 516 features and ~10.000 rows.

Dependencies

  • scikit-learn
  • pandas
  • numpy
  • matplotlib
  • seaborn
  • tensorflow
  • factor_analyzer
  • timeit
  • imblearn

File Description

  • input (Contain project dataset)
  • Project.ipynb (Main notebook for RFE model)
  • Neural Network.ippynb (Neural Network model train)
  • Grouped-activites.ipynb (Notebook for dataset grouped into 4 activities)

About

CS3244 Group 26 project - Human Activity and Postural Transition Recognition

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •