Loading and visualizing sensor data

After collecting sensor data we need to read the data into running software to work with it. The pandas toolkit contains tools for reading in data.

recording = pd.read_csv(recording_path) see

As a first step visualizing the data is important to understand what we’ve collected. Matplotlib can be used to visualize data in line plots.

a[x][y].plot(col_data) see

Line plots of gesture linear accelerometer recording.

Histograms can also give us information about the data.

Histograms of gesture linear accelerometer recording.

Running many quick data checks can be accomplished with small test scripts.

gesture_data = gesture_recording[gesture]
        column_histograms(gesture_data, name=f"{gesture} gesture", filepath=os.path.join(test_output, f"{gesture}-histograms.png"))
        plot_columns(gesture_data, name=f"{gesture} gesture", filepath=os.path.join(test_output, f"{gesture}-plots.png"))
see

The plots show the time is very different than the accelerometer readings constantly increasing in the line plots and evenly distributed in the histogram. The mean of the aT data is much higher than the other values. Using values of greatly different ranges in training an AI model could overweight some column measure over others. We need to balance the measure by normalizing column data into similar ranges before using the data for model training. Visualizing data lets us understand and sanity check the data before using it to train models (See notebook).