The steps involved in this recipe are as follows:
- Prep the data with prepdata.py:
import pandas as pd
df = pd.read_csv('Beach_Water_Quality_-_Automated_Sensors.csv',
header=0)
df = df[df['Beach Name'] == 'Rainbow Beach']
df = df[df['Water Temperature'] > -100]
df = df[df['Wave Period'] > -100]
df['Measurement Timestamp'] = pd.to_datetime(df['Measurement
Timestamp'])
Turbidity = df[['Measurement Timestamp', 'Turbidity']]
Turbidity.to_csv('Turbidity.csv', index=False, header=False)
- Import libraries in Luminol.py:
from luminol.anomaly_detector import AnomalyDetector
import time
- Perform anomaly detection:
my_detector = AnomalyDetector('Turbidity.csv')
score = my_detector.get_all_scores()
- Print the anomalies:
for (timestamp, value) in score.iteritems():
t_str = time.strftime('%y-%m-%d %H...