↑ ↓ Navigate ↵ Select ESC Close

Anomaly detection in time series with Python | Data Science with Marco

2023-04-16 Science & Technology
62.4k
1.2k
40
Data Science with Marco
Data Science with Marco
6.0k subscribers

Hands-on Anomaly Detection in Time Series Data using Python Methods

Learn three distinct Python methods—Robust Z-score, Isolation Forest, and Local Outlier Factor—to accurately identify rare, deviating events in your time series data.

Short Summary

  • Understand why anomaly detection is crucial for monitoring system health and ensuring stable forecasting models.
  • Implement a baseline Robust Z-score method tailored for non-normal data using the Median Absolute Deviation (MAD).
  • Evaluate tree-based (Isolation Forest) and density-based (LOF) algorithms, comparing their success rates on real AWS CPU utilization data.
  • The session moves between theoretical explanation and immediate code implementation in a Jupyter Notebook environment.

This lesson details the theoretical foundations and practical Python implementation for finding outliers in time series. Readers gain insight into method selection based on data characteristics, demonstrated by testing three techniques against labeled real-world CPU utilization metrics.

Unlock all features

FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.

Get My 10 Free Today

Related videos

Your Brain is a Time Machine, with Dean Buonomano
49:15

Your Brain is a Time Machine, with Dean Buonomano

StarTalk

26.7k views

We Try Science Experiments Without Instructions
27:41

We Try Science Experiments Without Instructions

Good Mythical Morning

190.8k views

Master Python Requests In 15 Minutes. Call Any API
16:52

Master Python Requests In 15 Minutes. Call Any API

Tech With Tim

31.3k views

Build & Deploy a Python AI Agent in 20 Minutes
23:16

Build & Deploy a Python AI Agent in 20 Minutes

Tech With Tim

38.6k views

Build a Python AI Agent in 10 Minutes
10:32

Build a Python AI Agent in 10 Minutes

Tech With Tim

84.2k views

How to Write Production Python Code
29:09

How to Write Production Python Code

Tech With Tim

63.1k views

How to Build AI Agents in Python
14:56

How to Build AI Agents in Python

Tech With Tim

46.7k views

How to Build a Production-Ready RAG AI Agent in Python (Step-by-Step)
1:16:29

How to Build a Production-Ready RAG AI Agent in Python (Step-by-Step)

Tech With Tim

77.8k views

How to Build an Advanced AI Agent with Search (LangGraph, Python, Bright Data & More)
1:33:01

How to Build an Advanced AI Agent with Search (LangGraph, Python, Bright Data & More)

Tech With Tim

44.4k views

Learn Pandas in 30 Minutes - Python Pandas Tutorial
30:37

Learn Pandas in 30 Minutes - Python Pandas Tutorial

Tech With Tim

166.6k views

Description

A hands-on lesson on detecting outliers in time series data using Python. Full source code: https://github.com/marcopeix/youtube_tutorials/blob/main/YT_02_anomaly_detection_time_series.ipynb Dataset can be found here: https://github.com/numenta/NAB/blob/master/data/realAWSCloudwatch/ec2_cpu_utilization_24ae8d.csv Labels can be found here: https://github.com/numenta/NAB/blob/master/labels/combined_labels.json Chapters: Introduction - 0:00 Get the data - 4:11 Robust Z-score method - 9:08 Robust Z-score method (code) - 13:12 Isolation forest - 20:48 Isolation forest (code) - 22:33 Local outlier factor - 27:16 Local outlier factor (code) - 31:21 Thank you - 34:01

Top Comments (10)

@EngMAli-vk3nz 2023-07-15

Thanks for this We Hope to make Some One For MultiVariate Time Series Anomaly Detection

7
@ab3180 2026-04-17

Nice

0
@shgo 2025-05-09

Really nice intro to the subject, Marco! Thanks for the tutorial! I'll definitely recommend that to my students.

0
@joaovict007 2024-03-07

Very interesting content, thank you!

0
@Fatimazahraelmhedden 2025-10-18

thank you very much . from Morocco

0
@littlepigywigy 2024-03-05

nice and clear

0
@Binsoyaxie 2026-04-15

Hello, i want to create an ai in an mobile android app that is on device learning or realtime learning, like the owner trains the ai in realtime like how the owner swipes, app usage, and app transistions, then after the ai is 100% trained, the ai is ready for possbible intruder in the phone How should I do this?

0
@PoulamiSenapati-u8x 2024-02-09

Hello Marco, thank you so much for such a great video. Can you please make a video on anomaly detection for time series data using pycaret.

0
@hoanhvong 2024-01-27

🎉 thank you a lot

0
@wisecarverrr 2025-04-25

Great video, thank you!

0

Unlock the Data Inside
Turn Videos into Knowledge

  • Get FREE 10/day: transcripts, summaries, chats
  • Chat with videos, export text & PDF
  • $1 free API credit for RAG, chatbots & research
Try it free

Free forever plan • All features unlocked

App screenshot