Open cochaviz opened 1 year ago
[+] Slicing the entire provided dataset now takes approximately 3 seconds. Please note, slicing now happens in standard intervals; meaning the start of the window is not anymore the first packet in the window. This is because this happened to be too time-consuming. This approach reduces the running time and does not harm functionality.
[+] The window_id is stored as a new column (not anymore passed as a parameter)
[+] Calculation of the statistical measures of the entire provided dataset now takes approximately 11 minutes. Here, the statistical measures are computed based on Mac Address for each window.
[+] Additionally the method mode_mean_med_without_window
has been added to compute the statistical measures of the entire provided dataset based on Mac Address (neglecting the window)
[+] The statistical results are saved in a generated CSV file
However, both calculation methods of the statistical measures do not seem to reproduce the results obtained by Pinheiro, Bezerra et al. Further investigation is required.
Slicing into time windows is very slow, as is calculating the mode/mean/median.