Caching strategy

Caching stores the results of code chunks so they don’t have to be re-run every time you render the document.

When to use caching

• Heavy Computation: Chunks taking >5 seconds (e.g., raster processing, MCMC).
• Stable Data Sources: Large datasets that change infrequently.

Disabling caching for specific chunks

Explicitly set #| cache: false for:

• Stochasticity: Simulations or bootstraps without a fixed set.seed().
• Data Export: Chunks using write_csv(), ggsave(), or saveRDS().

Common pitfalls

• Hidden Dependencies: Use #| dependson: "chunk-A" if B relies on A.
• External Data: Use #| cache-extra: !expr file.mtime("data.csv").