telling the user to wait a bit

the models load whenever the first request is made (20-30 minutes) but subsequent requests will work much faster; can be fixed with a pro subscription.

app.py

@@ -1,17 +1,21 @@
 from flask import Flask, request, jsonify
 from flask_cors import CORS
-import sys
 import os
+import sys
 
 sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
-from api.predict import predict_review
+from api.predict import predict_review, models_loaded
 
 app = Flask(__name__)
 CORS(app)
 
 @app.route('/health', methods=['GET'])
 def health():
-    return jsonify({"status": "ok", "model": "ensemble-v1"}), 200
+    return jsonify({
+        "status": "ok",
+        "model": "ensemble-v1",
+        "models_loaded": models_loaded
+    }), 200
 
 @app.route('/predict', methods=['POST'])
 def predict():
@@ -29,6 +33,13 @@ def predict():
         if len(reviewText.strip()) == 0:
             return jsonify({"error": "text cannot be empty"}), 400
         
+        if not models_loaded:
+            return jsonify({
+                "status": "loading",
+                "message": "models are loading for the first time, this will take 20-30 minutes. please wait...",
+                "models_loaded": False
+            }), 202
+        
         result = predict_review(reviewText)
         
         if "error" in result and result["prediction"] == "error":
@@ -50,5 +61,5 @@ def predict():
 
 if __name__ == '__main__':
     print("starting ensemble api server", flush=True)
-    print("loading models on first request...", flush=True)
-    app.run(host='0.0.0.0', port=5000, debug=False)
+    print("models will load on first prediction request", flush=True)
+    app.run(host='0.0.0.0', port=7860, debug=False)