Metadata-Version: 2.1
Name: kcc2020-tutorial-HAR-dataset
Version: 0.0.6
Summary: UNKNOWN
Home-page: UNKNOWN
Author: Auk Kim
Author-email: kimauk@kaist.ac.kr
License: UNKNOWN
Description: # Dataset for KCC2020: Tutorial on Human Activity Recognition 
        
        ## Install
        ```cmd
        pip install kcc2020-tutorial-HAR-dataset
        ```
        ## How to use
        * This dataset is for the KCC2020 tutorial on human activity recognition.  
        * This dataset is originally from UCI's HAPT (Human Activities and Postural Transitions) an modified for KCC 2020 Sensing tutorial. 
        * You can download original dataset from the [HAPT website](http://archive.ics.uci.edu/ml/datasets/Smartphone-Based+Recognition+of+Human+Activities+and+Postural+Transitions) (for details of the dataset, see [HAPT dataset website](http://archive.ics.uci.edu/ml/datasets/Smartphone-Based+Recognition+of+Human+Activities+and+Postural+Transitions)).
        
        ### load_all
        * To download dataset of entire users, use 'load_all().' 
        
        * For example:
        ```python
        from KCC2020_HAR_dataset import load_all
        
        entire_datase = load_all()
        ```
        
        * It automatically removes any part of data that has no labels. 
        
        * To obtain data with the removal, please use 'load_all(remove_no_lavels = False).' 
        
        ### load_by_user
        * To download dataset of a specific user, use 'load_by_user(uid).' 
        
        * It returns pandas's DataFrame that contains the dataset of the specific user with given uid. 
        
        * There is 30 users; it returns 'None' if uid is greater than 30.
        
        * For example:
        ```python
        from KCC2020_HAR_dataset import load_by_user
        
        user1_datasett = load_by_user(1)
        ```
        
        * It automatically removes any part of data that has no labels. 
        
        * To obtain data with the removal, please use 'load_by_user(uid, remove_no_lavels = False).' 
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering
Requires-Python: >=3.6
Description-Content-Type: text/markdown
