Open-source alternatives to Ts2vec

Lag-llama is a probabilistic machine learning foundation model designed for time series forecasting. It generates predictive distributions and uncertainty bounds for sequential data across arbitrary frequencies by leveraging pre-trained foundational weights. The system supports zero-shot transfer inference, allowing it to predict future values on entirely new and unseen datasets without requiring prior retraining. It achieves this by combining generalized representations from foundational training with adjustable context lengths, where historical context lengths and lagged feature values feed

Practical PyTorch is a collection of deep learning tutorials and guides focused on implementing recurrent neural networks. The project provides practical code for building sequence models and sequence-to-sequence architectures using the PyTorch framework. The repository covers the implementation of models for neural machine translation, character-level text generation, and text classification. It includes examples for transforming input sequences into output sequences for machine translation and synthesizing new text. The project also extends to sequence data prediction and time series analy

Univariate GARCH models in R

Multivariate GARCH Models

Code for Generative Time-series Modeling with Fourier Flows.

The requirements.txt file are attached for list of packages required. Python 3.9.16 torch==2.0.0 scikitlearn==0.24.2 pywavelets==1.4.1 pandas scipy statsmodels matplotlib Bottleneck

A Python toolkit for rule-based/unsupervised anomaly detection in time series

GluonTS is a framework for probabilistic time series forecasting, designed to predict future values as probability distributions with confidence intervals. It supports both traditional model training and zero-shot forecasting, where pretrained models generate predictions for new series without additional training. The project distinguishes itself by integrating a wide variety of forecasting approaches into a unified workflow. This includes deep learning architectures such as recurrent neural networks and causal convolutions, as well as the integration of external statistical models, the Proph

ROCKET · MINIROCKET · HYDRA

The Tidymodels Extension for GARCH models

tsfresh is an automated feature engineering tool and library designed to extract statistical characteristics from raw time series data. It transforms sequential data into tabular datasets, converting time series into a flat format where each row represents a unique entity and columns represent extracted features. The project distinguishes itself through a parallel data processing framework that distributes heavy computational workloads across multiple CPU cores. It also implements hypothesis-based feature selection to identify the most predictive characteristics and filter out irrelevant ones

This repository includes the official project for the paper: TransMix: Attend to Mix for Vision Transformers, CVPR 2022

Paper published at ICML22. Link to our paper: https://icml.cc/virtual/2022/spotlight/18038

Time-aware tibbles

Bringing financial analysis to the tidyverse

Time series analysis in the tidyverse

A python library for time-series smoothing and outlier detection in a vectorized way.

Automatic discovery of non-trivial statistical truths from 500+ public time series — mutual information, Granger causality, FDR correction

by Mutian Xu, Runyu Ding, Hengshuang Zhao, and Xiaojuan Qi.

Tian Zhou, Peisong Niu, Xue Wang, Liang Sun, Rong Jin, "One Fits All: Power General Time Series Analysis by Pretrained LM,", NeurIPS, 2023. paper

:construction: WIP :construction: