GIZ/Development-Project-Synergy-Finder

1import faiss
2
3"""
4Semantic Search Function
5"""
6def search(query, model, embeddings, filtered_df, top_x=20):
7
8        filtered_df_indecies_list = filtered_df.index
9        filtered_embeddings = embeddings[filtered_df_indecies_list]
10        
11        # Load or create FAISS index
12        dimension = filtered_embeddings.shape[1]  
13        faiss_index = faiss.IndexFlatL2(dimension)
14        faiss_index.add(filtered_embeddings)
15
16        # Convert query to embedding
17        query_embedding = model.encode([query])[0].reshape(1, -1)
18        
19        # Perform search
20        D, I = faiss_index.search(query_embedding, k=top_x)  # Search for top x similar items
21        
22        # Extract the sentences corresponding to the top indices
23        top_indecies = [i for i in I[0]]
24
25        return filtered_df.iloc[top_indecies]