(1/2) Google released a new foundation model for time series forecasting 🚀
The TimeFM (Time Series Foundation Model) is a foundation model for time series forecasting applications. This pre-trained model was developed by the Google Research team. It joins the recent trend of leveraging foundation models for time series forecasting, which includes Salesforce's Moirai and Amazon's Chronos.
Version 0.12.0 of the skforecast Python library for time series forecasting with regression models was released this week. The release includes new features, updates for existing ones, and bug fixes. 🧵👇🏼
Version 1.7.1 of the NeuralForecast #Python library was released last month by Nixtla. The NeuralForecast library, as the name implies, provides a neural network framework for time series forecasting. 🧵👇🏼
(1/2) Moirai - Salesforce's Foundation Forecasting Model 🚀
Salesforce recently released Moirari - a new #Python 🐍 library with a foundation model for time series forecasting applications. According to the release blog - the model comes with universal forecasting capabilities and can handle multiple scenarios and different frequencies.
Yesterday, Amazon released a new open-source project, Chronos - a family of pre-trained time series forecasting models based on language model architectures.
If you are looking to learn the foundation of time series analysis and forecasting, I recommend checking the Applied Time Series Analysis course from the University of Washington. The course by Prof Eli Holmes, Prof. Eric Ward, and Prof. Mark Scheuerell focuses on the foundation of time series analysis with applications for environmental science.
Soo been obsessing over #math matrix & #graph theory
Thinking options for most efficient #tech application using concept of a merkle tree for a matrix; still requiring efficient merkle #matrix verification (tree=type of graph)
Saw merkle field= #hash each layer, merging into node
(1/3)Modeling Short Time Series with Prior Knowledge in PyMC 🚀
Yesterday, I shared an article by Tim Radtke about forecasting insufficient time series data with a Bayesian approach using R. Here is the Python version 🧵👇🏼
(1/2) Modeling Short Time Series with Prior Knowledge
When modeling time series data, you may find yourself with insufficient data. Insufficient data in time series would typically be defined as less than one seasonal cycle. This would challenge us to understand whether some events are driven by seasonality or other reasons, such as one-time events, outliers, etc.
(5/7)
This year, I also retired two major open-source projects 👋🏼:
➡️ TSstudio - my first open-source project ❤️, R package for descriptive and predictive analysis of time series data 👇🏼
🔗 https://github.com/RamiKrispin/TSstudio
➡️ Coronavirus - R package provides a tidy format for the COVID-19 dataset collected by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University.
🔗 https://github.com/RamiKrispin/coronavirus
A nice survey paper (led by @bendfulcher) about ways to measure similarity/distance between a pair of multivariate #TimeSeries. It explains an organising structure around the different bits of literature that compares time series.
The timetk, one of the main R packages for time series analysis and forecasting ❤️, by Matt Dancho, is now available in Python 🐍. The package provides a variety of tools for working with time series data and analyzing it. The Python version leverages pandas for processing time series data and plotly for visualization.
Acaba de ser publicado o artigo "Multi-output time series forecasting with randomized multivariate Fuzzy Cognitive Maps" no periódico international Chaos, Solitons, and Fractals. Esse é um trabalho de autoria de Omid Orang, minha e de Frederico Gadelha e apresenta de um novo método baseado em redes neurais aleatorizadas (Reservoir Computing) e conjuntos fuzzy para previsão multivariada de séries temporais.
I spent last night to build this fun Shinylive app - Forecasting Sandbox 😎
The app provides a simple sandbox for three simple forecasting models - Linear regression, ARIMA, and Holt-Winters, and it entirely runs on the browser!
I am planning to deploy it to Github Actions and create a tutorial (WIP) 👇🏼
🤯 Holy sh*t, #sympy is such a crazy powerful :python: #Python package, look how much you can get out of a single one-line expression and some fancy decorators and wrappers.
The TSstudio package is one of the first #opensource projects I did. On my way to holiday break on December 2017, I was bored during a 12 hours flight and decided to package a few common #dataviz functions I love to use for time series data. A few days after, the first version was on CRAN. In the following year, I spent many hours enjoying developing additional functionalities and tools for #timeseries analysis and #forecasting
(1/2) After returning last week to work on my book, this weekend continue to work on the last dataset I will package for the book - the hourly electricity generation in New York by sub-regions.
Introduction to forecasting with ARIMA and seasonal ARIMA with Python 🚀👇🏼
This short tutorial by Joaquín Amat Rodrigo and Javier Escobar Ortiz provides an introduction to forecasting with ARIMA models using Python 🐍. This includes using different flavors of ARIMA methods from the statsmodels, pmdarima, skforecast, and statsForecast libraries.
The skforecast Python 🐍 library provides ML applications for time series forecasting using different regression models from the scikit-learn library. Here is a tutorial by Joaquín Amat Rodrigo and Javier Escobar Ortiz for time series forecasting with the skforecast using XGBoost, LightGBM, Scikit-learn, and CatBoost models 🚀.
If you need to brush the dust off your way of using a db query language, then try #PromQL. It has some similarities to things you know but also is handled differently. #prometheus#timeseries