Pipeline for Epidemic Modling

In this section, we will walk you through the building of epidemic modeling tasks using Epilearn, from dataset construction to model evaluation.

If you haven’t installed or tried Epilearn, please refer to Installation and Quickstart.

Dataset construction

Internal 2. External 3. Simulation

UniversalDataset

Transformations

supported transformations 2. how it works 3. customizing your own transformations

transformation = transforms.Compose({
                                    "features": [transforms.normalize_feat()],
                                    "graph": [transforms.normalize_adj()]})
dataset.transforms = transformation

Model Customization

Model categories 2. Internal Models 3. customized Models

Tasks Initialization

supported tasks and functions 2. how it works

task = Forecast(prototype=STGCN,
                dataset=None,
                lookback=lookback,
                horizon=horizon,
                device='cpu')

Evaluation

training and testing 2. tuning configurations 3. visualizing results