Sync up tweaks/changes

README.md

@@ -66,7 +66,7 @@ the libraries, model weights, and services upon which it is built.
 ## Getting Started
 
 To use this demo, you'll need accounts set up on [Google Cloud](https://cloud.google.com),
-[Modal](https://www.modal.com), and the Copernicus Data Store](<https://cds.climate.copernicus.eu/>). 
+[Modal](https://www.modal.com), and the [Copernicus Data Store](<https://cds.climate.copernicus.eu/>).
 Don't worry - even though you do need to supply them ith credit card information, this
 demo should cost virtually nothing to run; we'll use very limited storage on Google
 Cloud Storage for forecast model outputs that we generate (a few cents per month if you

ai-models-modal/main.py

@@ -109,6 +109,8 @@ def __init__(
             assets=config.AI_MODEL_ASSETS_DIR,
             date=int(self.init_datetime.strftime("%Y%m%d")),
             time=self.init_datetime.hour,
+            # TODO: allow user to specify desired forecast lead time, with limited
+            # validation (e.g. < 10 days)
             lead_time=12,
             path=str(self.out_pth),
             metadata={},  # Read by the output data handler
@@ -140,7 +142,9 @@ def generate_forecast(
     ai_model = AIModel(model_name, init_datetime)
 
     logger.info("Generating forecast...")
-    # ai_model.run_model.remote()
+    ai_model.run_model.remote()
+    # TODO: Modal re-runs the __init__ method after this call for some reason, look into
+    # why that's the case and if we can avoid it.
     logger.info("Done!")
 
     # Double check that we successfully produced a model output file.