stumpy

#STUMPY is a powerful and scalable open source Python library for modern time series analysis. Follow us to get the latest updates!

[stumpy @stumpy_dev]

Hello, time series people! 👋

This is the official account for .@stumpy_dev, a powerful and scalable Python package for modern time series analysis. 📈Check out this video to get started: https://t.co/ostFYPCHOP

Follow us for the latest news about STUMPY! 🗞

[Sean Law 🇨🇦 @seanmylaw]

🔥 My recent .@SciPyConf 2024 talk on ".@stumpy_dev Modern Time Series Analysis with Matrix Profiles" is now available! 🙏 Please share with your network 🚀

https://t.co/BEM8gqXvwV

[Sean Law 🇨🇦 @seanmylaw]

🔥 STUMPY 1.13.0 has been released!

Highlights:
- 🚀 Easier-to-Use Matrix Profile (Array) Data Structure
- 💪 NumPy 2.0 Support
- ❤️ pyproject.toml Adoption
- 😱 Improved Docs and Testing
- 🐍 Python 3.12 Support
- 🎢 And Much More!

Find out more: https://t.co/2BSOTsTGET

[Sean Law 🇨🇦 @seanmylaw]

😍 I'm super excited to be back at SciPy this year where I'll be presenting a quick intro to "modern time series analysis with matrix profiles" using STUMPY! If you are also attending the conference please say, "hello" 👋

[NumFOCUS @NumFOCUS]

A huge welcome to @stumpy_dev as a new NumFOCUS Affiliated Project! STUMPY is a powerful & scalable Python package for modern time series analysis & works to provide scientific researchers & data scientists with scalable & high-performance software for computing matrix profiles.

[stumpy @stumpy_dev]

@awesome_panel @Panel_org @seanmylaw You can learn more about STUMPY here: https://t.co/WC1R83W79S

[awesome-panel.org @awesome_panel]

#Stumpy is a powerful and scalable Python library for identifying repeating patterns in your timeseries via a matrix profile.

👉Check out this @Panel_org Stumpy app https://t.co/8i2ZeDtJj0

#python #dataviz #datascience #Analytics #timeseries @seanmylaw

[Sean Law 🇨🇦 @seanmylaw]

@marktenenholtz Thank you for your kind words .@marktenenholtz! .@stumpy_dev has been a labor of love and feedback like this (especially on the tutorials) means a lot!

[Mark Tenenholtz @marktenenholtz]

STUMPY is one of the most powerful time-series analysis methods (and my personal favorite)... and it quite literally is just a distance vector.

I would say every library should have an explainer written this clearly, but not every method is so elegant and simple! https://t.co/HYalWK50OS

[Sean Law 🇨🇦 @seanmylaw]

👉 This long-awaited .@stumpy_dev release (for modern time series analysis) has finally dropped and is packed full of goodies and even more of the high quality docs that you’ve come to love and expect! 😍

📣 Special thanks to .@sarajpoor for all of his many, many contributions!

[stumpy @stumpy_dev]

🔥 STUMPY 1.12.0 has been released!

Highlights:
- 🚀15-20% speed boost
- 💪Efficient top-k nearest-neighbors
- ❤️Better handling of constant subsequences
- 😱.@numba_jit function caching
- 🐍#python 3.11 support
- 🎢And much more!

📄Find out more: https://t.co/BR5dNIMRZy

[stumpy @stumpy_dev]

@retiagod @seanmylaw @predict_addict Fixed! Thank you for pointing it out @retiagod

[Valeriy M., PhD, MBA, CQF @predict_addict]

You would not use spoon 🥄 as a fork 🍴 or a knife. Why would data scientists try to use forecasting to solve anomaly detection problems. This is so wrong.

Anomaly detection is an own domain that has its own methods.

One of the pitfalls of #datascientists is trying to treat #anomalydetection as a forecasting problem and use forecasting models to try to detect anomalies using forecasting methods. This is wrong, an example below is a simple illustration of why such methods don’t work, in a simple time series two artificial anomalies were injected dropping signal value to zero.

For the first anomaly, the value of zero is still within “prediction interval” (which is most likely to be wrong as well unless conformal prediction has been used but this is not the subject of this post).

Forecasting based model then sees the value of zero as within prediction interval and clearly won’t label 🏷️ this anomaly as anomaly that it clearly is.

Don’t treat anomaly detection problems as forecasting problems, learn about anomaly detection as its own domain and what works and what does not work there.

Follow Eamonn Keogh works as he is #1 Professor globally in the field of anomaly detection.

Look at libraries like STUMPY created by @seanmylaw to deal with anomaly detection problems.

Learn about how #conformalprediction has successfully tackled anomaly detection (conformal prediction powers Microsoft Azure anomaly detection module) and other methods that work.

You would not use spoon 🥄 as a fork 🍴 or a knife.

@seanmylaw has got you covered, the best series of articles on matrix profile (research by Eamonn Keogh) explaining the matrix profile methods in plain English

https://t.co/49qNx7gguu

#anomalydetection #machinelearning #timeseries

[Sean Law 🇨🇦 @seanmylaw]

@predict_addict Thanks .@predict_addict! You can learn more about .@stumpy_dev by visiting:

https://t.co/YKyfoNlcbP

[stumpy @stumpy_dev]

@marktenenholtz https://t.co/WC1R83W79S

[Mark Tenenholtz @marktenenholtz]

In fact, methods like STUMPY exist for this.

Unlike the deep, 768/1536-dimensional embedding representations you need in NLP, "matrix profile" methods like STUMPY are pretty low-dimensional.

In other words: simpler models for simpler representations.

Now, the practicalities:

[Marc Skov Madsen, PhD, CFA® @MarcSkovMadsen]

Check out the #Stumpy @bokeh app by @seanmylaw

App: https://t.co/sl97QyFZoO
Code: https://t.co/8W2A9johi2

[Marc Skov Madsen, PhD, CFA® @MarcSkovMadsen]

Did you know the hvPlot and the @HoloViz_org ecosystem can help you do well in @kaggle competitions?

Check out "Horse Racing - Welcome to the Machine" notebook that won silver https://t.co/D7crhzG45X

@Panel_org #dataviz #dataanalysis #datascience #analytics #HorseRacing

[Marc Skov Madsen, PhD, CFA® @MarcSkovMadsen]

For example, you can share @bokeh apps like the #Stumpy app by @seanmylaw.

App: https://t.co/sl97QyFrzg
Code:https://t.co/8W2A9jnJsu

🧵👇

[Federico Carrone @federicocarrone]

@seanmylaw @stumpy_dev I've been following the development and it's really amazing. Congrats!

[Sean Law 🇨🇦 @seanmylaw]

🙏 Thanks .@federicocarrone! We’ve been working hard over the past year to expand core .@stumpy_dev features, significantly improve the speed/performance, and increasing the stability of the codebase (without sacrificing 100% code coverage). We’re excited so please stay tuned! 👀