Rolling window for inference (#335)

* adding make_Rt methods

* more make Rt methods

* Make endemic scenarios geometric mean 1

* Update test_constructors.jl

* prefix methods

* format fix

* Update infer.jl

* add prefix for simulate

* define a `lookback` parameter

* Update make_tspan.jl

* Default params for stride of rolling window

* new make_tspan and modify other areas of code that assume a fixed tspan

pipeline/src/constructors/make_default_params.jl

@@ -22,12 +22,16 @@ function make_default_params(pipeline::AbstractEpiAwarePipeline)
     α_delay = 4.0
     θ_delay = 5.0 / 4.0
     lookahead = 21
+    lookback = 35
+    stride = 7
     return Dict(
         "Rt" => Rt,
         "logit_daily_ascertainment" => logit_daily_ascertainment,
         "cluster_factor" => cluster_factor,
         "I0" => I0,
         "α_delay" => α_delay,
         "θ_delay" => θ_delay,
-        "lookahead" => lookahead)
+        "lookahead" => lookahead,
+        "lookback" => lookback,
+        "stride" => stride)
 end

pipeline/src/constructors/make_inference_configs.jl

@@ -8,18 +8,20 @@ Create inference configurations for the given pipeline. This is the default meth
 - An object representing the inference configurations.
 
 """
-function make_inference_configs(pipeline::AbstractEpiAwarePipeline)
+function make_inference_configs(pipeline::AbstractEpiAwarePipeline; start = 21)
     gi_param_dict = make_gi_params(pipeline)
     namemodel_vect = make_epiaware_name_latentmodel_pairs(pipeline)
     igps = make_inf_generating_processes(pipeline)
     obs = make_observation_model(pipeline)
     priors = make_model_priors(pipeline)
     default_params = make_default_params(pipeline)
+    N = size(make_Rt(pipeline), 1)
+    Ts = start:default_params["stride"]:N |> collect
 
     inference_configs = Dict("igp" => igps, "latent_namemodels" => namemodel_vect,
         "observation_model" => obs, "gi_mean" => gi_param_dict["gi_means"],
         "gi_std" => gi_param_dict["gi_stds"], "log_I0_prior" => priors["log_I0_prior"],
-        "lookahead" => default_params["lookahead"]) |>
+        "lookahead" => default_params["lookahead"], "lookback" => default_params["lookback"], "T" => Ts) |>
                         dict_list
 
     selected_inference_configs = _selector(inference_configs, pipeline)

pipeline/src/constructors/make_tspan.jl

@@ -1,16 +1,23 @@
 """
-Constructs the time span for the given `pipeline` object.
+Constructs a time span for performing inference on a case data time series. This
+    is the default method.
 
 # Arguments
-- `pipeline::AbstractEpiAwarePipeline`: The pipeline object for which the time
-    span is constructed. This is the default method.
+- `pipeline::AbstractEpiAwarePipeline`: The pipeline object used for analysis.
+- `T::Union{Integer,Nothing} = nothing`: The `stop` point at which to construct
+    the time span. If `nothing`, the time span will be constructed using the
+    length of the Rt vector for `pipeline`.
+- `lookback = 35`: The number of days to look back from the specified time point.
 
 # Returns
-- `tspan::Tuple{Float64, Float64}`: The time span as a tuple of start and end times.
+A tuple `(start, stop)` representing the start and stop indices of the time span.
+
+# Examples
 
 """
-function make_tspan(pipeline::AbstractEpiAwarePipeline; backhorizon = 21)
+function make_tspan(pipeline::AbstractEpiAwarePipeline;
+        T::Union{Integer, Nothing} = nothing, lookback = 35)
     N = size(make_Rt(pipeline), 1)
-    @assert backhorizon<N "Backhorizon must be less than the length of the default Rt."
-    return (1, N - backhorizon)
+    _T = isnothing(T) ? N : T
+    return (max(1, _T - lookback), min(N, _T))
 end

pipeline/src/infer/generate_inference_results.jl

@@ -15,7 +15,9 @@ Generate inference results based on the given configuration of inference model o
 """
 function generate_inference_results(
         truthdata, inference_config, pipeline::AbstractEpiAwarePipeline;
-        tspan, inference_method, datadir_name = "epiaware_observables")
+        inference_method, datadir_name = "epiaware_observables")
+    tspan = make_tspan(
+        pipeline; T = inference_config["T"], lookback = inference_config["lookback"])
     config = InferenceConfig(
         inference_config; case_data = truthdata["y_t"], tspan, epimethod = inference_method)
 
@@ -46,8 +48,9 @@ which is deleted after the function call.
 - `inference_results`: The generated inference results.
 """
 function generate_inference_results(
-        truthdata, inference_config, pipeline::EpiAwareExamplePipeline;
-        tspan, inference_method)
+        truthdata, inference_config, pipeline::EpiAwareExamplePipeline; inference_method)
+    tspan = make_tspan(
+        pipeline; T = inference_config["T"], lookback = inference_config["lookback"])
     config = InferenceConfig(inference_config; case_data = truthdata["y_t"],
         tspan = tspan, epimethod = inference_method)
 
@@ -65,8 +68,9 @@ end
 Method for prior predictive modelling.
 """
 function generate_inference_results(
-        inference_config, pipeline::RtwithoutRenewalPriorPipeline;
-        tspan)
+        inference_config, pipeline::RtwithoutRenewalPriorPipeline)
+    tspan = make_tspan(
+        pipeline; T = inference_config["T"], lookback = inference_config["lookback"])
     config = InferenceConfig(
         inference_config; case_data = missing, tspan, epimethod = DirectSample())
 

pipeline/src/infer/map_inference_results.jl

@@ -15,9 +15,9 @@ tasks from `Dagger.@spawn`.
 
 """
 function map_inference_results(
-        truthdata, inference_configs, pipeline::AbstractEpiAwarePipeline; tspan, inference_method)
+        truthdata, inference_configs, pipeline::AbstractEpiAwarePipeline; inference_method)
     map(inference_configs) do inference_config
         Dagger.@spawn generate_inference_results(
-            truthdata, inference_config, pipeline; tspan, inference_method)
+            truthdata, inference_config, pipeline; inference_method)
     end
 end

pipeline/src/pipeline/do_inference.jl

@@ -11,9 +11,8 @@ An array of inference results.
 """
 function do_inference(truthdata, pipeline::AbstractEpiAwarePipeline)
     inference_configs = make_inference_configs(pipeline)
-    tspan = make_tspan(pipeline)
     inference_method = make_inference_method(pipeline)
     inference_results = map_inference_results(
-        truthdata, inference_configs, pipeline; tspan, inference_method)
+        truthdata, inference_configs, pipeline; inference_method)
     return inference_results
 end

pipeline/test/constructors/test_constructors.jl

@@ -145,7 +145,9 @@ end
         "I0" => 100.0,
         "α_delay" => 4.0,
         "θ_delay" => 5.0 / 4.0,
-        "lookahead" => 21
+        "lookahead" => 21,
+        "lookback" => 35,
+        "stride" => 7
     )
 
     # Test the make_default_params function