This is perfectly timed! I am currently working on my masters thesis where i have to analyse the energy consumption of households, which is time series data. Thank you so much!
@@patrick-gerard Moin, ich hab gesehen das du auch deutsch bist :D klar, bräuchte unterstützung bei neuronalen netzen, wenn du dich damit auskennst? Wie könnten wir denn connecten? :)
Does anyone else face this error while plotting? Looks like matplotlib does not have this style in use anymore. OSError: 'seaborn-whitegrid' is not a valid package style, path of style file, URL of style file, or library style name (library styles are listed in `style.available`)
Hi Mate, Thank you very much for this great video. I am trying to find something similar to the anomaly models you showed in the video but for multivariate analysis time series and if it is possible find the most one contributing to the change in the main time series?. Can you please help me with that.
It was excellent, but it has not been maintained since April 2020. For example, I had an issue with an old pandas dependency. I think it was on ThresholdAD. it's a shame.
🎯 Key Takeaways for quick navigation: 01:02 📈 *The Anomaly Detection Toolkit (ADTK) in Python offers rule-based time series anomaly detection without relying on machine learning models.* 01:42 🌐 *Anomaly detection is valuable for various applications, such as server monitoring for potential attacks or identifying unusual stock price movements for informed trading decisions.* 03:03 📊 *The video demonstrates the use of ADTK with time series data sets, including global temperature data and stock prices obtained through the Yahoo Finance Library.* 06:19 📉 *Various anomaly detection methods, including threshold, quantile, interquartile range, and generalized extreme studentized deviate (ESD) tests, are explained and demonstrated.* 12:24 📉 *The tutorial also covers persist anomaly detection, which compares each value to the previous one, helping identify positive or negative changes in time series data, demonstrated using stock price data, especially useful for detecting unusual price movements.* Made with HARPA AI