Tutorials
Runnable tutorials across the EpiAware packages.
Every tutorial below is a runnable walkthrough that lives in its package’s own documentation, where it is executed and kept in step with the code. Filter by the approach it belongs to, or search across all of them. If you are new to an approach, start with the tutorial marked start here.
| Tutorial | What it covers |
|---|---|
| Composing distributions start here | The core walkthrough. Build an event tree from named delays, then simulate and fit with the same object. ComposedDistributions |
| Linear chain | The simplest tree, with delays chained in series from onset through to outcome. ComposedDistributions |
| Competing outcomes | Where a case's path forks, death or discharge, as a fixed-probability mixture or as racing hazards. ComposedDistributions |
| Strata and uncertainty | Delays that vary across groups, and delays whose parameters are themselves uncertain. ComposedDistributions |
| Concepts | Maps each modelling idea to the verb that builds it. ComposedDistributions |
| Overview start here | How the three roles, prior, infection, and observation, nest through one interface. ComposableTuringIDModels |
| Renewal with negative-binomial observations | A renewal infection process fitted to case counts with overdispersed noise. ComposableTuringIDModels |
| Real-time nowcast | Estimating what has happened but not yet been reported, from right-truncated data. ComposableTuringIDModels |
| Delays and day-of-week effects | Layering reporting delays and weekday reporting patterns onto the observation model. ComposableTuringIDModels |
| Split observations | Fitting one infection process to more than one observed data stream at once. ComposableTuringIDModels |
| SIR ODE model | Compartmental dynamics as the infection process, written directly. ComposableTuringIDModels |
| Catalyst ODE model | The same compartmental dynamics, built through Catalyst. ComposableTuringIDModels |
| Design | The composable design set out in full. ComposableTuringIDModels |
| Fitting with Turing | Estimating a delay distribution from censored, truncated line-list data. CensoredDistributions |
| Analytical primary-censored CDFs | Closed-form primary-event censoring, and when it beats numerical integration. CensoredDistributions |
| Exponentially tilted primary events | Primary events that are not uniform within their window, as in a growing epidemic. CensoredDistributions |
| Automatic differentiation backends | Which AD backend to use for censored delay likelihoods, and what it costs. CensoredDistributions |
| Composed chains | Modifier leaves used inside a composed event tree, the packages working together. ModifiedDistributions |
| Modifier pipeline | Rescaling, shifting, and transforming a delay while keeping the distribution interface. ModifiedDistributions |
| Weighted likelihoods | Weighting an observation's likelihood, the standard trick for aggregated or count data. ModifiedDistributions |
| Visualising convolutions | Adding delays together, and what the resulting distribution looks like. ConvolvedDistributions |
No tutorials match that search.
Have a tutorial you would like to see? Suggest it on the forum or see Contributing. Analyses built with these tools are in the Gallery.