Create embedding.py

src/embedding.py

@@ -0,0 +1,20 @@
+# embedding.py
+import os
+import numpy as np
+import pandas as pd
+import faiss
+from sentence_transformers import SentenceTransformer
+
+# --- Load data ---
+def load_data():
+    data_path = os.path.join(os.path.dirname(__file__), 'train_data.csv')
+    df = pd.read_csv(data_path)
+    return df['question'].tolist(), df['answer'].tolist()
+
+# --- Embedding model and FAISS index ---
+def setup_embeddings(answers):
+    embedder = SentenceTransformer('sentence-transformers/paraphrase-MiniLM-L6-v2')
+    answer_embeddings = embedder.encode(answers, show_progress_bar=True)
+    index = faiss.IndexFlatL2(answer_embeddings.shape[1])
+    index.add(np.array(answer_embeddings))
+    return embedder, index