search

geekan / MetaGPT

🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming
https://deepwisdom.ai/
MIT License
45.64k stars 5.43k forks source link

ValueError: Creator not registered for key: LLMType.OLLAMA #1553

Open vanhocpham opened 1 month ago

vanhocpham commented 1 month ago

Bug description

I using MetaGPT ver 0.8.1 but when use RAG with method SimpleEngine.from_docs have error ValueError: Creator not registered for key: LLMType.OLLAMA

Environment information

Screenshots or logs

config2.yaml embedding: api_type: "ollama" model: "hf.co/hugging-quants/Llama-3.2-3B-Instruct-Q8_0-GGUF" base_url: "http://127.0.0.1:11434/api"

llm: api_type: "ollama" model: "hf.co/hugging-quants/Llama-3.2-3B-Instruct-Q8_0-GGUF" base_url: "http://127.0.0.1:11434/api"

Error Response /usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/base.py in get_instance(self, key, kwargs) 27 return creator(kwargs) 28 ---> 29 raise ValueError(f"Creator not registered for key: {key}") 30 31

ValueError: Creator not registered for key: LLMType.OLLAMA

better629 commented 1 month ago

@vanhocpham try the main branch, it supports ollama in https://github.com/geekan/MetaGPT/blob/main/metagpt/rag/factories/embedding.py#L26

vanhocpham commented 1 month ago

@better629 i using pip install is working?

better629 commented 1 month ago

@vanhocpham refs to https://docs.deepwisdom.ai/main/en/guide/get_started/installation.html#install-in-development-mode

cd MetaGPT
pip3 install -e .
vanhocpham commented 1 month ago

@better629 Thanks, I will try and get back to you

vanhocpham commented 1 month ago

@better629 I tried the way you instructed but another error occurred 

/usr/local/lib/python3.10/dist-packages/metagpt/rag/engines/simple.py in from_docs(cls, input_dir, input_files, transformations, embed_model, llm, retriever_configs, ranker_configs)
    116         nodes = run_transformations(documents, transformations=transformations)
    117 
--> 118         return cls._from_nodes(
    119             nodes=nodes,
    120             transformations=transformations,

/usr/local/lib/python3.10/dist-packages/metagpt/rag/engines/simple.py in _from_nodes(cls, nodes, transformations, embed_model, llm, retriever_configs, ranker_configs)
    230         llm = llm or get_rag_llm()
    231 
--> 232         retriever = get_retriever(configs=retriever_configs, nodes=nodes, embed_model=embed_model)
    233         rankers = get_rankers(configs=ranker_configs, llm=llm)  # Default []
    234 

/usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/retriever.py in get_retriever(self, configs, **kwargs)
     72             return self._create_default(**kwargs)
     73 
---> 74         retrievers = super().get_instances(configs, **kwargs)
     75 
     76         return SimpleHybridRetriever(*retrievers) if len(retrievers) > 1 else retrievers[0]

/usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/base.py in get_instances(self, keys, **kwargs)
     16     def get_instances(self, keys: list[Any], **kwargs) -> list[Any]:
     17         """Get instances by keys."""
---> 18         return [self.get_instance(key, **kwargs) for key in keys]
     19 
     20     def get_instance(self, key: Any, **kwargs) -> Any:

/usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/base.py in <listcomp>(.0)
     16     def get_instances(self, keys: list[Any], **kwargs) -> list[Any]:
     17         """Get instances by keys."""
---> 18         return [self.get_instance(key, **kwargs) for key in keys]
     19 
     20     def get_instance(self, key: Any, **kwargs) -> Any:

/usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/base.py in get_instance(self, key, **kwargs)
     44         creator = self._creators.get(type(key))
     45         if creator:
---> 46             return creator(key, **kwargs)
     47 
     48         self._raise_for_key(key)

/usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/retriever.py in _create_faiss_retriever(self, config, **kwargs)
     87 
     88     def _create_faiss_retriever(self, config: FAISSRetrieverConfig, **kwargs) -> FAISSRetriever:
---> 89         config.index = self._build_faiss_index(config, **kwargs)
     90 
     91         return FAISSRetriever(**config.model_dump())

/usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/retriever.py in wrapper(self, config, **kwargs)
     45         if index is not None:
     46             return index
---> 47         return build_index_func(self, config, **kwargs)
     48 
     49     return wrapper

/usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/retriever.py in _build_faiss_index(self, config, **kwargs)
    128         vector_store = FaissVectorStore(faiss_index=faiss.IndexFlatL2(config.dimensions))
    129 
--> 130         return self._build_index_from_vector_store(config, vector_store, **kwargs)
    131 
    132     @get_or_build_index

/usr/local/lib/python3.10/dist-packages/metagpt/rag/factories/retriever.py in _build_index_from_vector_store(self, config, vector_store, **kwargs)
    156     ) -> VectorStoreIndex:
    157         storage_context = StorageContext.from_defaults(vector_store=vector_store)
--> 158         index = VectorStoreIndex(
    159             nodes=self._extract_nodes(config, **kwargs),
    160             storage_context=storage_context,

/usr/local/lib/python3.10/dist-packages/llama_index/core/indices/vector_store/base.py in __init__(self, nodes, use_async, store_nodes_override, embed_model, insert_batch_size, objects, index_struct, storage_context, callback_manager, transformations, show_progress, **kwargs)
     74 
     75         self._insert_batch_size = insert_batch_size
---> 76         super().__init__(
     77             nodes=nodes,
     78             index_struct=index_struct,

/usr/local/lib/python3.10/dist-packages/llama_index/core/indices/base.py in __init__(self, nodes, objects, index_struct, storage_context, callback_manager, transformations, show_progress, **kwargs)
     75             if index_struct is None:
     76                 nodes = nodes or []
---> 77                 index_struct = self.build_index_from_nodes(
     78                     nodes + objects,  # type: ignore
     79                     **kwargs,  # type: ignore

/usr/local/lib/python3.10/dist-packages/llama_index/core/indices/vector_store/base.py in build_index_from_nodes(self, nodes, **insert_kwargs)
    308             print("Some nodes are missing content, skipping them...")
    309 
--> 310         return self._build_index_from_nodes(content_nodes, **insert_kwargs)
    311 
    312     def _insert(self, nodes: Sequence[BaseNode], **insert_kwargs: Any) -> None:

/usr/local/lib/python3.10/dist-packages/llama_index/core/indices/vector_store/base.py in _build_index_from_nodes(self, nodes, **insert_kwargs)
    277             run_async_tasks(tasks)
    278         else:
--> 279             self._add_nodes_to_index(
    280                 index_struct,
    281                 nodes,

/usr/local/lib/python3.10/dist-packages/llama_index/core/indices/vector_store/base.py in _add_nodes_to_index(self, index_struct, nodes, show_progress, **insert_kwargs)
    230 
    231         for nodes_batch in iter_batch(nodes, self._insert_batch_size):
--> 232             nodes_batch = self._get_node_with_embedding(nodes_batch, show_progress)
    233             new_ids = self._vector_store.add(nodes_batch, **insert_kwargs)
    234 

/usr/local/lib/python3.10/dist-packages/llama_index/core/indices/vector_store/base.py in _get_node_with_embedding(self, nodes, show_progress)
    137 
    138         """
--> 139         id_to_embed_map = embed_nodes(
    140             nodes, self._embed_model, show_progress=show_progress
    141         )

/usr/local/lib/python3.10/dist-packages/llama_index/core/indices/utils.py in embed_nodes(nodes, embed_model, show_progress)
    136             id_to_embed_map[node.node_id] = node.embedding
    137 
--> 138     new_embeddings = embed_model.get_text_embedding_batch(
    139         texts_to_embed, show_progress=show_progress
    140     )

/usr/local/lib/python3.10/dist-packages/llama_index/core/instrumentation/dispatcher.py in wrapper(func, instance, args, kwargs)
    309 
    310             try:
--> 311                 result = func(*args, **kwargs)
    312                 if isinstance(result, asyncio.Future):
    313                     # If the result is a Future, wrap it

/usr/local/lib/python3.10/dist-packages/llama_index/core/base/embeddings/base.py in get_text_embedding_batch(self, texts, show_progress, **kwargs)
    333                     payload={EventPayload.SERIALIZED: self.to_dict()},
    334                 ) as event:
--> 335                     embeddings = self._get_text_embeddings(cur_batch)
    336                     result_embeddings.extend(embeddings)
    337                     event.on_end(

/usr/local/lib/python3.10/dist-packages/llama_index/embeddings/ollama/base.py in _get_text_embeddings(self, texts)
     73         embeddings_list: List[List[float]] = []
     74         for text in texts:
---> 75             embeddings = self.get_general_text_embedding(text)
     76             embeddings_list.append(embeddings)
     77 

/usr/local/lib/python3.10/dist-packages/llama_index/embeddings/ollama/base.py in get_general_text_embedding(self, texts)
     86     def get_general_text_embedding(self, texts: str) -> List[float]:
     87         """Get Ollama embedding."""
---> 88         result = self._client.embeddings(
     89             model=self.model_name, prompt=texts, options=self.ollama_additional_kwargs
     90         )

/usr/local/lib/python3.10/dist-packages/ollama/_client.py in embeddings(self, model, prompt, options, keep_alive)
    279     keep_alive: Optional[Union[float, str]] = None,
    280   ) -> Mapping[str, Sequence[float]]:
--> 281     return self._request(
    282       'POST',
    283       '/api/embeddings',

/usr/local/lib/python3.10/dist-packages/ollama/_client.py in _request(self, method, url, **kwargs)
     73       response.raise_for_status()
     74     except httpx.HTTPStatusError as e:
---> 75       raise ResponseError(e.response.text, e.response.status_code) from None
     76 
     77     return response

ResponseError: 404 page not found
seehi commented 1 month ago

Try removing "/api" from the base_url of the embedding, such as

embedding:
  api_type: "ollama"
  model: "hf.co/hugging-quants/Llama-3.2-3B-Instruct-Q8_0-GGUF"
  base_url: "http://127.0.0.1:11434"
github-actions[bot] commented 2 hours ago

This issue has no activity in the past 30 days. Please comment on the issue if you have anything to add.