Human Activity Recognition (HAR)

Contents

The utils folder contains the code for reading and processing the data into a tensor form. The generated tensors have the dimensions

(batch, seq_len, n_channels)

where batch is the number of training examples in each batch, seq_len is the number of steps in the time series (128) and n_channels is the number of channels where observations are made (9).

The UCIHAR folder is our homemade dataset, which follows the same data processing used in the publicly released UCIHAR dataset, and it contains 1150 observations: 1000 of which are used to train the model, and the remaining ones for testing purposes.

The aim is to classify the activities correctly, which are

1 RESTING
2 STANDING
3 WALKING
4 FALLING
5 WALKING_UPSTAIRS
6 WALKING_DOWNSTAIRS
7 STANDING_UP
8 SITTING

Result

Method	Test accuracy
CNN	94%

CNN architecture