Deep learning is perhaps the nearest future of human activity recognition. While there are many existing non-deep method, we still want to unleash the full power of deep learning. This repo provides a demo of using deep learning to perform human activity recognition.
We support both Tensorflow and Pytorch.
- Python 3.x
- Numpy
- Tensorflow or Pytorch 1.0+
There are many public datasets for human activity recognition. You can refer to this survey article Deep learning for sensor-based activity recognition: a survey to find more.
In this demo, we will use UCI HAR dataset as an example. This dataset can be found in here.
Of course, this dataset needs further preprocessing before being put into the network. I've also provided a preprocessing version of the dataset as a .npz file so you can focus on the network (download HERE). It is also highly recommended you download the dataset so that you can experience all the process on your own.
| #subject | #activity | Frequency |
|---|---|---|
| 30 | 6 | 50 Hz |
-
For Pytorch (recommend), go to
pytorchfolder, config the folder of your data inconfig.py', and then runmain_pytorch.py`. -
For tensorflow, run
main_tensorflow.pyfile. The update of tensorflow version is stopped since I personally like Pytorch.
What is the most influential deep structure? CNN it is. So we'll use CNN in our demo.
Convolution + pooling + convolution + pooling + dense + dense + dense + output
That is: 2 convolutions, 2 poolings, and 3 fully connected layers.
That dataset contains 9 channels of the inputs: (acc_body, acc_total and acc_gyro) on x-y-z. So the input channel is 9.
Dataset providers have clipped the dataset using sliding window, so every 128 in .txt can be considered as an input. In real life, you need to first clipped the input using sliding window.
So in the end, we reformatted the inputs from 9 inputs files to 1 file, the shape of that file is [n_sample,128,9], that is, every windows has 9 channels with each channel has length 128. When feeding it to Tensorflow, it has to be reshaped to [n_sample,9,1,128] as we expect there is 128 X 1 signals for every channel.