Deep learning PyTorch library for time series forecasting

Time series Forecasting on the NTU_RGB+D skeleton dataset using AutoFormer and FEDFormer

PyTorch image classification models pre-trained

A novel deep transfer learning framework for forecasting NavIC satellite ephemeris and clock errors. PRESTO overcomes extreme data scarcity using a hybrid architecture (GNNs + Semiparametric Decomposition + Autoformers) and synthetic data augmentation to generate normally distributed prediction residuals.

Training a financial forecasting model based on the AutoFormer architecture with PyTorch and Alpaca.py

A novel deep transfer learning framework for forecasting NavIC satellite ephemeris and clock errors. PRESTO overcomes extreme data scarcity using a hybrid architecture (GNNs + Semiparametric Decomposition + Autoformers) and synthetic data augmentation to generate normally distributed prediction residuals.

autoformer-forecasting

An autoregressive forecasting implementation of a LSTM network, NBEATS architecture, ARIMA and SARIMAX regressions, and Autoformer architecture on rupee dollar exchange rates using pytorch, pytorch lightning, pytorch-forecasting, and GluonTS

Tokenization Matters: A Fair Ablation of Point-wise, and Variate-wise Transformers for Financial Time Series. (includes PatchTST, iTransformer, Crossformer, Autoformer, Fedformer, Informer, TimeNet, and non-stationarity extensions.

This study examines the effectiveness of transformer-based models for financial time series forecasting, specifically focusing on log returns derived from daily closing prices of the DAX40 index. We propose a decoder-only transformer model designed for immediate-term financial time series forecasting: The PatternDecoder.