Sensitivity Dispatcher

The sensitivity dispatcher is a tool for running one or more predefined sensitivity experiments using the dataset notebooks/double-entry-data/double_entry_final.csv. It is intended to use when running experiments on a server.

See also

Documentation for Sensitivity Dispatcher Examples

Documentation for Individual Sensitivity Experiments

Sensitivity Dispatcher Usage

usage: python scripts/sensitivity_dispatcher.py [-h]
                                                [--max_processes MAX_PROCESSES]
                                                [--categories CATEGORIES [CATEGORIES ...]]
                                                [--dry_run]
                                                [--model_type MODEL_TYPE]

Named Arguments

--max_processes

Number of processes to spawn

--categories

Run types to execute

--dry_run

Print run types selected and exit

Default: False

--model_type

Model type to use for requested sensitivity analyses

Default: “default”

Sensitivity Analysis Dispatcher Run Types

Full specifications can be found in scripts/sensitivity_analysis/sensitivity_analysis.yaml, which can also be customised to your needs.

Sensitivity Dispatcher Run Types

Run Type	Description	Experiment File
region_holdout	Mask data for one region and output prediction for the masked region. Runs once for each region.	region_holdout.py
npi_leaveout	Remove NPI indicators from dataset and fit model with the remaining set of NPIs. Runs once for each NPI, and additionally leaves out school and university closures jointly	npi_leaveout.py
cases_threshold	Run the model masking daily confirmed cases when a region’s total number of confirmed cases is below [10, 50, 150, 200] cases	preprocessing_tests.py
deaths_threshold	Run the model masking daily death counts when a region’s total number of deaths is below [1, 5, 30, 50] deaths	preprocessing_tests.py
oxcgrt	Run the model with additional features from OxCGRT: [‘Travel Screen/Quarantine’, ‘Travel Bans’, ‘Public Transport Limited’, ‘Internal Movement Limited’, ‘Public Information Campaigns’, ‘Symptomatic Testing’]	oxcgrt_leavein.py
R_prior	Run the model with a prior mean R0 of [2.28, 2.78, 3.78, 4.38]	alternative_build_param.py
growth_noise	Add noise to growth rate	alternative_build_param.py
npi_prior	Run the model with an NPI effetiveness prior of [skewed 10, Normal(0,0.2), ICL]	alternative_build_param.py
agg_holdout	Mask the final 20 days of data and predict this period with the model	agg_holdout.py
structural	Run the model with alternative structures: [additive, discrete_renewal_fixed_gi, noisy_r, different_effects, cases_only, deaths_only]
epiparam	Run the model with alternative priors over generation interval, infection to death timing and infection to case recording	epiparam.py
iceswe	Run the model holdout out both iceland and sweden	iceswe.py
scaling	Run the model scaling daily numbers of cases	scaling.py