Neural Network Sensordata as Input

I have a dataset consisting of sensor recordings about human movement. There are 22 classes of different movement like sitting or walking and 19 sensor values.

Each recording of a movement has about 1000 lines contained in a csv file.

My problem: I don't know how to present those recordings to a neural network (TensorFlow) so that it can be trained on the movement classes and even predict what was done in recording by getting those 19000 values. I don't even know which Neural Network Model I should use and therefore need your help.

Topic tensorflow python bigdata

Category Data Science

1
answered at

Correct me if I'm wrong: As far as I understood, you have files of 1000 entries with 19 numerical sensor values and a type of 22 movements in each line. You would like to predict the type of movement based on the sensor values.

Are the entries being recorded in a fixed interval? If so, you could use an RNN.

Either way, it seems to be a classification problem. You can find a tutorial for Tensorflow here.

You can read data from csv files with pandas.read_csv.

As for the encoding of your data: If the types of movement cannot be sorted and are not related to each other, you should use one-hot encoding to present the type of movement to the NN. You can keep numerical sensor data as is, just make sure to normalize it.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.