Update app.py

app.py

@@ -11,24 +11,18 @@ import numpy as np
 model = SentenceTransformer('all-MiniLM-L6-v2')
 sentence_embed = pd.read_csv('Reference_file.csv')
 
-#st.write(sentence_embed.head(5))
-
-# Function to compute cosine similarity
-def cosine_similarity(v1, v2):
-    """Compute cosine similarity between two vectors."""
-    return np.dot(v1, v2) / (np.linalg.norm(v1) * np.linalg.norm(v2))
+import streamlit as st
 
-# Backend function for mapping
+# Define the function for mapping code
 def mapping_code(user_input):
-    emb1 = model.encode(user_input)
+    emb1 = model.encode(user_input.lower())
     similarities = []
-    for sentence_emb in sentence_embed['embeds']:
-        sentence_emb = np.array(sentence_emb)
-        similarity = cosine_similarity(sentence_emb, emb1)
+    for sentence_embed in sentences['embeds']:
+        similarity = util.cos_sim(sentence_embed, emb1)
         similarities.append(similarity)
 
     # Combine similarity scores with 'code' and 'description'
-    result = list(zip(sentence_embed['SBS Code'], sentence_embed['Long Description'], similarities))
+    result = list(zip(sentences['SBS Code'], sentences['Long Description'], similarities))
 
     # Sort results by similarity scores
     result.sort(key=lambda x: x[2], reverse=True)
@@ -40,7 +34,6 @@ def mapping_code(user_input):
         top_5_results.append({"Code": code, "Description": description, "Similarity Score": similarity_score})
     return top_5_results
 
-
 # Streamlit frontend interface
 def main():
     st.title("CPT Description Mapping")
@@ -62,4 +55,4 @@ def main():
                 st.write(f"{i}. Code: {result['Code']}, Description: {result['Description']}, Similarity Score: {result['Similarity Score']:.4f}")
 
 if __name__ == "__main__":
-    main()
+    main()