Data synchronization

linmod/data.py

@@ -26,6 +26,7 @@
 import sys
 from datetime import datetime
 from pathlib import Path
+from typing import Optional
 from urllib.parse import urlparse
 from urllib.request import urlopen
 
@@ -133,14 +134,11 @@
 """
 
 
-if __name__ == "__main__":
-    # Load configuration, if given
-
+def main(cfg: Optional[dict]):
     config = DEFAULT_CONFIG
 
-    if len(sys.argv) > 1:
-        with open(sys.argv[1]) as f:
-            config["data"] |= yaml.safe_load(f)["data"]
+    if cfg is not None:
+        config["data"] |= cfg["data"]
 
     # Download the data, if necessary
 
@@ -267,3 +265,14 @@
     model_df.write_csv(config["data"]["save_path"]["model"])
 
     print_message(" done.")
+
+
+if __name__ == "__main__":
+    # Load configuration, if given
+
+    cfg = None
+    if len(sys.argv) > 1:
+        with open(sys.argv[1]) as f:
+            cfg = yaml.safe_load(f)["data"]
+
+    main(cfg)

present-day-forecasting/README.md

@@ -1,6 +1,10 @@
-# Demo of model fitting and evaluation
+# Model fitting and evaluation
 
-Current workflow, from top-level of repo:
+From top-level of repo, run `present-day-forecasting/main.py present-day-forecasting/config.yaml`.
 
-1. Download data with `python3 -m linmod.data present-day-forecasting/config.yaml`
-2. Fit and evaluate models with `present-day-forecasting/main.py present-day-forecasting/config.yaml`
+This:
+1. Creates datasets for forecasting and evaluation (from a cached dataset if available).
+2. Runs all forecasting models specified in `present-day-forecasting/config.yaml` and plots a simple summary.
+3. Evaluates the model forecasts using the created evaluation dataset.
+
+Parameters failing MCMC convergence diagnostics are reported, but not excluded from downstream steps.

present-day-forecasting/main.py

@@ -10,6 +10,7 @@
 import yaml
 from numpyro.infer import MCMC, NUTS
 
+import linmod.data
 import linmod.eval
 import linmod.models
 from linmod.utils import print_message
@@ -27,6 +28,9 @@
 with open(sys.argv[1]) as f:
     config = yaml.safe_load(f)
 
+# Create the datasets
+linmod.data.main(config)
+
 # Load the dataset used for retrospective forecasting
 
 data = pl.read_csv(config["data"]["save_file"]["model"], try_parse_dates=True)