def do_analysis(): riffdata = Riffdata() meetings = riffdata.get_meetings_with_participant_utterances() print(f'Found utterances from {len(meetings)} meetings') meeting_ids = list(meetings) # Just to get an idea of what is in a meeting w/ participant utterances print 1 _print_participant_uts_info(meeting_ids[0], meetings[meeting_ids[0]]) durations = get_utterance_durations(meetings) print(f'Found {len(durations)} utterances') durations.sort() print( f'shortest utterance was {durations[0]}ms and longest was {durations[-1]}ms' ) buckets, bucket_cnt, graph_ranges = _distribute_durations(durations) _print_bucket_data(buckets, bucket_cnt) x, y = _make_xy_sets_to_plot(x_src=buckets, y_src=bucket_cnt, ranges=graph_ranges) # pylint: disable=consider-using-enumerate fig, ax = plt.subplots(len(x), 1) for plot in range(0, len(x)): my_plotter(ax[plot], x[plot], y[plot], {'marker': 'x'}) fig.savefig('plot.png')
def do_analysis(): riffdata = Riffdata() meetings = riffdata.get_meetings_with_participant_utterances() print(f'Found utterances from {len(meetings)} meetings') gaps = get_utterance_gaps(meetings) x = np.array(gaps) fig, ax = plt.subplots() # the histogram of the data (see example: https://matplotlib.org/gallery/statistics/histogram_features.html) num_bins = 50 ax.hist(x, num_bins, range=(0, 4000)) fig.savefig('plot_gap.png')
def do_analysis(): riffdata = Riffdata() meetings = riffdata.get_meetings_with_participant_utterances() print(f'Found utterances from {len(meetings)} meetings') zerolen_ut_distribution = get_zerolen_ut_distribution(meetings) percentile_size = 100 // len(zerolen_ut_distribution) x = np.array(range(percentile_size, 101, percentile_size)) y = np.array(zerolen_ut_distribution) # pylint: disable=consider-using-enumerate fig, ax = plt.subplots() my_plotter(ax, x, y, {'marker': 'x'}) fig.savefig('plot_0_distrib.png')