fix config usage and remove email field

docs/source/schema.rst

@@ -32,7 +32,6 @@ The schema is as follows:
                 "postal_code": "M5E1 B07",
                 "country": "Canada"
             },
-            "email": "[email protected]",
             "metadata": {
                 "type": "Medical Clinic",
                 "languages": [
@@ -59,5 +58,4 @@ The schema is a JSON object with the following fields:
     - ``province``: The province where the service is located.
     - ``postal_code``: The postal code of the service location.
     - ``country``: The country where the service is located.
-- ``email``: The email address of the service.
 - ``metadata``: Additional metadata about the service. The metadata object can have any number of fields.

health_rec/api/data.py

@@ -101,8 +101,6 @@ class Service(BaseModel):
         A list of phone numbers associated with the service.
     address : Address
         The physical address of the service.
-    email : str
-        The email address of the service.
     metadata : Dict[str, Any]
         Additional metadata associated with the service.
     last_updated : Optional[datetime]
@@ -117,7 +115,6 @@ class Service(BaseModel):
     longitude: float
     phone_numbers: List[PhoneNumber]
     address: Address
-    email: str
 
     # Optional metadata fields stored as key-value pairs
     metadata: Dict[str, Any] = Field(default_factory=dict)

health_rec/services/rag.py

@@ -79,7 +79,7 @@ def _generate_response(self, query: str, context: str) -> RecommendationResponse
 
         try:
             completion = self.client.chat.completions.create(
-                model="gpt-4o",
+                model=Config.OPENAI_MODEL,
                 messages=[
                     {
                         "role": "user",

health_rec/services/refine.py

@@ -124,7 +124,7 @@ def improve_query(
             """
 
             completion = self.client.chat.completions.create(
-                model="gpt-4o",
+                model=Config.OPENAI_MODEL,
                 messages=[{"role": "user", "content": prompt}],
                 max_tokens=300,
             )

health_rec/services/rerank.py

@@ -84,7 +84,7 @@ def rerank(
         try:
             messages = self._create_ranking_prompt(query, services)
             completion = self.client.chat.completions.create(
-                model="gpt-4o",
+                model=Config.OPENAI_MODEL,
                 messages=messages,  # type: ignore
                 temperature=0,
             )

scripts/download_211_data.py

@@ -1,5 +1,6 @@
 """Download and process data from the 211 API."""
 
+import argparse
 import json
 import logging
 import math
@@ -175,8 +176,6 @@ def main() -> None:
     """Main function to run the script."""
     load_dotenv("./.env.development")
 
-    import argparse
-
     parser = argparse.ArgumentParser(description="Download data from the 211 API.")
     parser.add_argument(
         "--api-key", default=os.getenv("211_API_KEY"), help="API key for 211 API"
@@ -186,7 +185,12 @@ def main() -> None:
         default="https://data.211support.org/api/v2/search",
         help="Base URL for 211 API",
     )
-    parser.add_argument("--dataset", default="on", help="Dataset to download")
+    parser.add_argument(
+        "--dataset",
+        default="211CX",
+        help="Dataset to download",
+        choices=["211CX", "on", "nl", "nb"],
+    )
     parser.add_argument(
         "--is-gta", action="store_true", help="Whether to download GTA data"
     )