Set up | Find the shops and our walkshed | What each block can reach | How many cafes each block can reach | Which cafe is closest

Key terms | Travel times | Build a client | Look things up instead of guessing | Make a call and map it | Choose an output | The client methods | Handle errors

Set up | See what is around | Find the blocks that qualify | Narrow to a shared commute

Set up | Pull the blocks, with population | Coverage, one amenity at a time | The 15-minute-city score | Who can reach all six | Which amenity to add first | Who is one or two amenities away | Site a new supermarket

Set up | Read travel times from a starting point | Map the supermarkets within a 30-minute walk | Draw how far you can walk | Walk versus transit

Free queries | The big levers | Isochrones are the cheap reach primitive | Watch what you spend

1: MBG model basics | 1.A: Input data | 1.B: ID raster | 2: Model family and link function | 3: Model effects | 3.A: Covariate effects | 3.A.i: Standard covariate effect | 3.A.ii: Stacked ensemble model | 3.B: Gaussian process | 3.C: Administrative-level effect | 3.D: Nugget | 4: Aggregation to polygon boundaries | 5: Logging | 6: Further reading

Setting up | Building MBG model inputs | Fit the geostatistical model | Plot gridded estimates | Plot aggregated estimates

Predictive validity metrics | Log predictive density | Watanabe-Aikake information criterion | Root mean squared error | Model cross-validation | Setup | In-sample model comparison | Out-of-sample model comparison | Summary | Further reading

Setup | Running ML models with mbg and caret | Stacking ML models