nth-attempt

``` from pyculiarity import detect_ts import pandas as pd twitter_example_data = pd.read_csv('raw_data.csv', usecols=['timestamp', 'count']) results = detect_ts(twitter_example_data, max_anoms=0.02, direction='both') plt.plot(twitter_example_data['timestamp'], twitter_example_data['value']) plt.plot(results['anoms'].anoms, 'o') ```

If I understand this correct, this approach will return the smaller sequence lengths first even with `shuffle=True`. That may not the the ideal behavior?