LABDA Accelerometers

A package designed to process data from movement sensors – accelerometers.

• Auto-calibration
• Non-wear detection
• Metrics: Counts, ENMO
• Python

See documentation for more details.

Installation

Install using pip install labda-accelerometers.

A Simple Example

import pandas as pd
from labda_accelerometers import Metrics, AutoCalibrate, WearDetection

df = AutoCalibrate().calibrate(df)
print(df)
#>                                         acc_x     acc_y     acc_z
#> datetime  
#> 2021-09-09 00:00:07.009999990+02:00 -0.099318 -0.128671  0.995101
#> 2021-09-09 00:00:07.019999981+02:00  0.076385 -0.267248  0.995101
#> 2021-09-09 00:00:07.029999971+02:00  0.092358 -0.267248  0.927356

epoch = 1 # In seconds

acc_wear = WearDetection(epoch=epoch).from_acceleration(df)
metrics = Metrics(epoch=epoch)

enmo = metrics.enmo(df)
counts = metrics.counts(df)

results = pd.concat([acc_wear, enmo, counts], axis=1)
print(results)
#>                             wear      enmo  counts_x  counts_y  counts_z  counts_vm
#> datetime  
#> 2021-09-09 00:00:07+02:00  False  0.022882         0         5        51  51.244511  
#> 2021-09-09 00:00:08+02:00  False  0.024908         0         0         6   6.000000  
#> 2021-09-09 00:00:09+02:00  False  0.014403         0         0         0   0.000000  

Detailed information on labda-accelerometers processing and features is available here.