Chore: Added console logs for debugging.

app.py

@@ -1,24 +1,18 @@
-# import sys
+import os
+import warnings
 import json
-
 import numpy as np
-from sentence_transformers import SentenceTransformer
+from typing import List
 
-import os
-import warnings
-# from fastapi import FastAPI, HTTPException
-# from pydantic import BaseModel
+from sentence_transformers import SentenceTransformer
 import gradio as gr
-from typing import List
+
 
 # Suppress warnings
 os.environ['TRANSFORMERS_VERBOSITY'] = 'error'
 os.environ['TOKENIZERS_PARALLELISM'] = 'false'
 warnings.filterwarnings("ignore")
 
-# Initialize FastAPI
-# app = FastAPI(title="SelahSearch NLP Agent")
-
 # Load Model
 MODEL_NAME = "odunola/sentence-transformers-bible-reference-final"
 model = SentenceTransformer(MODEL_NAME)
@@ -31,6 +25,7 @@ THEMES = ["Trust and Guidance", "Restoration and Peace", "Wrath and Judgment", "
 THEME_VECS = model.encode(THEMES, convert_to_numpy=True)
 THEME_VECS = THEME_VECS / np.linalg.norm(THEME_VECS, axis=1, keepdims=True)
 
+
 def chunk_text(text):
     words = text.split()
     return [" ".join(words[i : i + 400]) for i in range(0, len(words), 200)]
@@ -47,15 +42,6 @@ def get_thematic_signature(doc_vec):
     norm = np.linalg.norm(relu_scores)
     return relu_scores / (norm if norm > 0 else 1.0)
 
-# # Define Data Models
-# class Song(BaseModel):
-#     name: str
-#     lyrics: str
-
-# class AnalysisRequest(BaseModel):
-#     passage: str
-#     songs: List[Song]
-
 
 def analyze_thematic_similarity(passage_text, songs_json):
     """
@@ -97,6 +83,7 @@ def analyze_thematic_similarity(passage_text, songs_json):
         # Sort by score descending
         results.sort(key=lambda x: x['score'], reverse=True)
         return results
+    
     except Exception as e:
         raise gr.Error(f"NLP Worker Error: {str(e)}")
     
@@ -114,51 +101,3 @@ demo = gr.Interface(
 
 if __name__ == "__main__":
     demo.queue().launch() # demo.queue() is vital for high RAM reliability
-
-
-# @app.get("/healthcheck")
-# def read_root():
-#     return {"Status": "Alive"}
-
-# @app.post("/analyse")
-# async def analyze_similarity(data: AnalysisRequest):
-#     try:
-#         p_vec = get_normalized_vector(data.passage)
-#         p_sig = get_thematic_signature(p_vec)
-
-#         results = []
-
-#         for song in data.songs:
-#             # Process Song Lyrics
-#             l_vec = get_normalized_vector(song.lyrics)
-#             direct_sim = float(np.dot(p_vec, l_vec))
-            
-#             relevant_themes = []
-#             final_score = direct_sim
-
-#             # Original Threshold Logic
-#             if direct_sim >= 0.1: 
-#                 l_sig = get_thematic_signature(l_vec)
-#                 thematic_sim = float(np.dot(p_sig, l_sig))
-                
-#                 contributions = p_sig * l_sig
-#                 relevant_themes = [THEMES[i] for i, val in enumerate(contributions) if val > 0.05]
-                
-#                 # 60/40 weighted split
-#                 final_score = (0.6 * direct_sim) + (0.4 * thematic_sim)
-
-#             results.append({
-#                 "name": song.name,
-#                 "score": round(final_score, 4),
-#                 "themes": relevant_themes
-#             })
-
-#         # Sort by score descending
-#         results.sort(key=lambda x: x['score'], reverse=True)
-#         return results
-#     except Exception as e:
-#         raise HTTPException(status_code=500, detail=str(e))
-
-# if __name__ == "__main__":
-#     import uvicorn
-#     uvicorn.run(app, host="0.0.0.0", port=int(os.getenv("PORT", 8000)))