RAG

app.py

@@ -17,11 +17,17 @@ def load_and_retrieve_docs(url):
     splits = text_splitter.split_documents(docs)
     embedding_model = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')
     
-    # Create a custom embedding function that uses the embedding model's encode method
-    def embed_func(texts):
-        return embedding_model.encode(texts, convert_to_tensor=True)
+    # Define a custom embedding function compatible with Chroma's interface
+    class CustomEmbeddings:
+        def __init__(self, model):
+            self.model = model
+
+        def embed_documents(self, texts):
+            return self.model.encode(texts, convert_to_tensor=True).tolist()
+
+    embeddings = CustomEmbeddings(embedding_model)
     
-    vectorstore = Chroma.from_documents(documents=splits, embedding=embed_func)
+    vectorstore = Chroma.from_documents(documents=splits, embedding=embeddings)
     return vectorstore.as_retriever()
 
 # Function to format documents