Time Series Forecasting with GRUs

Overview

This course explores Gated Recurrent Units (GRUs) in PyTorch for multivariate time series forecasting. We will build, evaluate, and apply advanced GRU techniques like Bidirectional GRUs, Attention Mechanisms, and Hybrid GRU-CNN models to improve forecasting accuracy.

Syllabus

Unit 1: Introduction to Time Series Forecasting with GRUs using PyTorch

Experimenting with GRU Architecture Parameters
Completing GRU Forecasting Model with Linear Output Layer
Adding Loss Function and Optimizer to GRU Model for Air Quality Forecasting
Training a GRU Model for Air Quality Temperature Prediction

Unit 2: Evaluating GRU Model Performance with PyTorch

Implementing Root Mean Squared Error (RMSE) Calculation for GRU Model Evaluation
Implementing Additional Evaluation Metrics for GRU Model
Implementing Adjusted R² Score for Time Series Model Evaluation
Visualize GRU Model Training and Validation Loss Curves
Visualizing Predicted vs Actual Values in Time Series Forecasting
Comparing GRU and SimpleRNN Models for Time Series Forecasting