still modfiying

Author: franciscojavierarceo

module_4_rag/README.md

@@ -21,8 +21,10 @@ Then run
 ```bash
 python batch_score_documents.py
 ```
-Which will output data to `city_wikipedia_summaries_with_embeddings.parquet`
+Which will output data to `data/city_wikipedia_summaries_with_embeddings.parquet`
 
+Next we'll need to do some Feast work and move the data into a repo created by
+Feast.
 ## Feast
 
 To get started, make sure to have Feast installed and PostGreSQL.

module_4_rag/app.py

@@ -6,30 +6,13 @@
 )
 from flasgger import Swagger
 from datetime import datetime
-from get_features import (
-    get_onboarding_features,
-    get_onboarding_score,
-    get_daily_features,
-    get_daily_score,
-)
-from ml import make_risk_decision
 
 app = Flask(__name__)
 swagger = Swagger(app)
 
 
-@app.route("/")
-def onboarding_page():
-    return render_template("index.html")
-
-
-@app.route("/home")
-def home_page():
-    return render_template("home.html")
-
-
-@app.route("/onboarding-risk-features/", methods=["POST"])
-def onboarding_features():
+@app.route("/get_documents")
+def get_documents():
     """Example endpoint returning features by id
     This is using docstrings for specifications.
     ---
@@ -39,60 +22,27 @@ def onboarding_features():
         in: query
         required: true
         default: NJ
-
-      - name: ssn
-        type: string
-        in: query
-        required: true
-        default: 123-45-6789
-
-      - name: dl
-        type: string
-        in: query
-        required: true
-        default: some-dl-number
-
-      - name: dob
-        type: string
-        in: query
-        required: true
-        default: 12-23-2000
     responses:
       200:
-        description: A JSON of features
+        description: A JSON of documents
         schema:
-          id: OnboardingFeatures
+          id: Document ID
           properties:
             is_gt_18_years_old:
               type: array
               items:
                 schema:
                   id: value
                   type: number
-            is_valid_state:
-              type: array
-              items:
-                schema:
-                  id: value
-                  type: number
-            is_previously_seen_ssn:
-              type: array
-              items:
-                schema:
-                  id: value
-                  type: number
-            is_previously_seen_dl:
-              type: array
-              items:
-                schema:
-                  id: value
-                  type: number
     """
-    r = request.args
-    feature_vector = get_onboarding_features(
-        r.get("state"), r.get("ssn"), r.get("dl"), r.get("dob")
-    )
-    return jsonify(feature_vector)
+    question = request.form["question"]
+    documents = store.get_online_documents(query)
+    return render_template("documents.html", documents=documents)
+
+
+@app.route("/")
+def home():
+    return render_template("home.html")
 
 
 if __name__ == "__main__":

module_4_rag/feature_repo/data/city_wikipedia_summaries.csv

@@ -10,7 +10,7 @@ online_store:
   host: 127.0.0.1
   port: 5432
   database: feast
-  user: ""
+  user: "demo_user"
   password: ""
 
 

module_4_rag/feature_repo/data/city_wikipedia_summaries_with_embeddings.parquet

@@ -9,7 +9,7 @@
 from entities import item
 
 
-parquet_file_path = "../data/city_wikipedia_summaries_with_embeddings.parquet"
+parquet_file_path = "./data/city_wikipedia_summaries_with_embeddings.parquet"
 
 source = FileSource(
     file_format=ParquetFormat(),