Pipelines Cleanup

pipelines/README.md

@@ -0,0 +1 @@
+# Pipelines

pipelines/fit_model.py → pipelines/fit_pyrenew_model.py

@@ -4,7 +4,9 @@
 
 import jax
 import numpy as np
-from build_model import build_model_from_dir
+from build_pyrenew_model import (
+    build_model_from_dir,
+)
 
 
 def fit_and_save_model(

pipelines/forecast_state.py

@@ -11,13 +11,15 @@
 import tomli_w
 import tomllib
 from prep_data import process_and_save_state
+from prep_eval_data import save_eval_data
 from pygit2 import Repository
-from save_eval_data import save_eval_data
 
 numpyro.set_host_device_count(4)
 
-from fit_model import fit_and_save_model  # noqa
-from generate_predictive import generate_and_save_predictions  # noqa
+from fit_pyrenew_model import fit_and_save_model  # noqa
+from generate_predictive import (
+    generate_and_save_predictions,
+)  # noqa
 
 
 def record_git_info(model_run_dir: Path):
@@ -125,13 +127,13 @@
     return None
 
 
-def postprocess_forecast(
+def plot_and_save_state_forecast(
     model_run_dir: Path, pyrenew_model_name: str, timeseries_model_name: str
 ) -> None:
     result = subprocess.run(
         [
             "Rscript",
-            "pipelines/postprocess_state_forecast.R",
+            "pipelines/plot_and_save_state_forecast.R",
             f"{model_run_dir}",
             "--pyrenew-model-name",
             f"{pyrenew_model_name}",
@@ -141,7 +143,7 @@
         capture_output=True,
     )
     if result.returncode != 0:
-        raise RuntimeError(f"postprocess_forecast: {result.stderr}")
+        raise RuntimeError(f"plot_and_save_state_forecast: {result.stderr}")
     return None
 
 
@@ -159,17 +161,17 @@
     return None
 
 
-def render_webpage(model_run_dir: Path) -> None:
+def render_diagnostic_report(model_run_dir: Path) -> None:
     result = subprocess.run(
         [
             "Rscript",
-            "pipelines/render_webpage.R",
+            "pipelines/diagnostic_report/render_diagnostic_report.R",
             f"{model_run_dir}",
         ],
         capture_output=True,
     )
     if result.returncode != 0:
-        raise RuntimeError(f"render_webpage: {result.stderr}")
+        raise RuntimeError(f"render_diagnostic_report: {result.stderr}")
     return None
 
 
@@ -363,11 +365,11 @@
     logger.info("Conversion complete.")
 
     logger.info("Postprocessing forecast...")
-    postprocess_forecast(model_run_dir, "pyrenew_e", "timeseries_e")
+    plot_and_save_state_forecast(model_run_dir, "pyrenew_e", "timeseries_e")
     logger.info("Postprocessing complete.")
 
     logger.info("Rendering webpage...")
-    render_webpage(model_run_dir)
+    render_diagnostic_report(model_run_dir)
     logger.info("Rendering complete.")
 
     if score:

pipelines/generate_predictive.py

@@ -3,7 +3,9 @@
 from pathlib import Path
 
 import arviz as az
-from build_model import build_model_from_dir
+from build_pyrenew_model import (
+    build_model_from_dir,
+)
 
 
 def generate_and_save_predictions(

pipelines/create_observed_data_tables.py → pipelines/hubverse_create_observed_data_tables.py

@@ -5,7 +5,10 @@
 
 import epiweeks
 import polars as pl
-from prep_data import aggregate_facility_level_nssp_to_state, get_state_pop_df
+from prep_data import (
+    aggregate_facility_level_nssp_to_state,
+    get_state_pop_df,
+)
 
 
 def save_observed_data_tables(

pipelines/save_eval_data.py → pipelines/prep_eval_data.py

@@ -3,7 +3,10 @@
 from pathlib import Path
 
 import polars as pl
-from prep_data import get_state_pop_df, process_state_level_data
+from prep_data import (
+    get_state_pop_df,
+    process_state_level_data,
+)
 
 
 def save_eval_data(