pytorch-forecasting

APDTFlow is a modern and extensible forecasting framework for time series data that leverages advanced techniques including neural ordinary differential equations (Neural ODEs), transformer-based components, and probabilistic modeling. Its modular design allows researchers and practitioners to experiment with multiple forecasting models and easily

Temporal Fusion Transformer model實作，目的為熟悉特徵工程、建模流程及預測結果視覺化，並深入研究模型架構與內部邏輯，強化對新模型的理解能力。當前仍在優化中

A deep learning-based application for predicting future stock prices using Temporal Fusion Transformer (TFT) and global market data.

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

End-to-end AI system for retail demand forecasting using Temporal Fusion Transformer (TFT) and dynamic pricing optimization using PPO reinforcement learning, combining prediction and decision-making in a unified pipeline

A plug and play framework for Temporal Fusion Transformer. Predict your future!

Comparative study of XGBoost and Temporal Fusion Transformers (TFT) for retail sales forecasting using feature engineering, time series cross-validation, and deep learning on the Rossmann Store Sales dataset.

PyTorch Forecasting SageMaker Examples

Multi-level retail demand forecasting system that enforces hierarchical coherence using optimal reconciliation methods. Combines bottom-up neural forecasts with top-down statistical models, then applies MinT reconciliation to ensure forecasts sum correctly across product/store hierarchies. Addresses the real-world challenge where independently trai

FinFusion: S&P 500 return forecasting with Temporal Fusion Transformers - compares TFT, ARIMAX, LSTM, and regime-aware variants.

Predicting institutional capture via Temporal Fusion Transformers. This data-centric project transforms 23 years of messy socioeconomic, fiscal, and electoral data into a mathematically balanced tensor to forecast "Fat" political dynasty saturation in the Philippines.

An end-to-end data engineering pipeline and multivariate time-series forecasting engine utilizing PyTorch Forecasting and Temporal Fusion Transformers for high-scale demand modeling.

7-day forecasting of the Dutch grid CO2 emission factor with Temporal Fusion Transformer

Compare feature-engineered XGBoost and Temporal Fusion Transformer models to forecast multi-store retail sales.

Design writeup, no code committed yet: planned TFT product trend forecaster fusing sales time series with social sentiment, with out-of-time validation and quantile forecasts.