Query the code memory. Args: query_text (str): The query text n_results (int): The number of results to return Returns: List[Dict]: The query results list
(self, query_text: str, n_results: int = 5)
| 67 | |
| 68 | |
| 69 | def query_code(self, query_text: str, n_results: int = 5) -> List[Dict]: |
| 70 | """ |
| 71 | Query the code memory. |
| 72 | |
| 73 | Args: |
| 74 | query_text (str): The query text |
| 75 | n_results (int): The number of results to return |
| 76 | |
| 77 | Returns: |
| 78 | List[Dict]: The query results list |
| 79 | """ |
| 80 | query_embedding = self.embedder.embeddings.create(input=[query_text], model="text-embedding-3-small").data[0].embedding |
| 81 | results = self.client.get_or_create_collection(self.collection_name).query(query_embeddings=[query_embedding], n_results=n_results) |
| 82 | return [ |
| 83 | { |
| 84 | "file": metadata['filenames'], |
| 85 | "content": doc |
| 86 | } |
| 87 | for doc, metadata in zip(results['documents'][0], results['metadatas'][0]) |
| 88 | ] |
| 89 | class DummyReranker(Reranker): |
| 90 | def __init__(self, model: str = None) -> None: |
| 91 | super().__init__(model) |