Upload 6 files

app.py

@@ -38,7 +38,7 @@ data = get_data()
 
 col1, col2 = st.columns([3, 3], gap="medium")
 
-instructor_model = get_instructor_embedding_model()
+#instructor_model = get_instructor_embedding_model()
 instructor_model_api = get_instructor_embedding_model_api()
 
 
@@ -105,7 +105,7 @@ dense_array = np.array(json_dict["data"], dtype=np.float64)
 dense_embedding_api = dense_array.tolist()
 
 
-dense_embedding = instructor_model.encode([[query_embedding_instruction, query_text]]).tolist()
+#dense_embedding = instructor_model.encode([[query_embedding_instruction, query_text]]).tolist()
 
 
 text_embedding_instructions_choice = [
@@ -144,7 +144,7 @@ with col2:
             pinecone_index = pinecone.Index(pinecone_index_name)
 
             matches = query_pinecone(
-                dense_vec=dense_embedding, top_k=num_results, index=pinecone_index, indices=indices
+                dense_vec=dense_embedding_api, top_k=num_results, index=pinecone_index, indices=indices
             )
             context = format_query(matches)
             output_text = format_context(context)

utils/retriever.py

@@ -30,7 +30,7 @@ def format_query(query_results):
 def format_context(context):
     output_text = []
     for text, score in context:
-        output_text.append(f"Text: {text}\nCosine Similarity: {score}")
+        output_text.append(f"Text: {text}\n\nCosine Similarity: {score}")
     return output_text