An error occurred: Error code: 400 - {'object': 'error', 'message': 'could not broadcast input array from shape (535,) into shape (512,)', 'type': 'BadRequestError', 'param': None, 'code': 400}

[Z-oo883]

There were no errors when building the knowledge base, but there were errors when querying. I use Qwen2.5-7BInstruct-GPTQ-Int4 as the large language model and bge-large-zh-v1.5 as the vector model. Use PDF file as input. please help me ! thank you !!

The code is as follows:

import os
import asyncio
from lightrag import LightRAG, QueryParam
from lightrag.llm import openai_complete_if_cache, openai_embedding
from lightrag.utils import EmbeddingFunc
import numpy as np
from lightrag.llm import hf_embedding
from transformers import AutoModel, AutoTokenizer

WORKING_DIR = "./dickens/"

if not os.path.exists(WORKING_DIR):
    os.mkdir(WORKING_DIR)

async def llm_model_func(
        prompt, system_prompt=None, history_messages=[], **kwargs
) -> str:
    return await openai_complete_if_cache(
        "Qwen/Qwen2.5-7B-Instruct-GPTQ-Int4",
        prompt,
        system_prompt=system_prompt,
        history_messages=history_messages,
        api_key=os.getenv("EMPTY"),
        base_url="http://0.0.0.0:8000/v1",
        **kwargs,
    )

async def main():
    try:
        rag = LightRAG(
            working_dir=WORKING_DIR,
            llm_model_func=llm_model_func,
            embedding_func=EmbeddingFunc(
                embedding_dim=1024,
                max_token_size=8192,
                func=lambda texts: hf_embedding(
                    texts,
                    tokenizer=AutoTokenizer.from_pretrained(
                        "bge-large-zh-v1.5", model_max_length=512
                    ),
                    embed_model=AutoModel.from_pretrained(
                        "bge-large-zh-v1.5"
                    ),
                ),
            ),
        )
        import textract

        file_path = '哈利波特第一章和第二章.pdf'
        text_content = textract.process(file_path)

        await rag.ainsert(text_content.decode('utf-8'))
        while(True):
            string = input()
            print(
            await rag.aquery(
                string,
                param=QueryParam(mode="hybrid"),
            )
        )
    except Exception as e:
        print(f"An error occurred: {e}")

if __name__ == "__main__":
    asyncio.run(main())

##error:
INFO:lightrag:Inserting 123 vectors to entities
We strongly recommend passing in an `attention_mask` since your input_ids may be padded. See https://huggingface.co/docs/transformers/troubleshooting#incorrect-output-when-padding-tokens-arent-masked.
You may ignore this warning if your `pad_token_id` (0) is identical to the `bos_token_id` (0), `eos_token_id` (2), or the `sep_token_id` (None), and your input is not padded.
INFO:lightrag:Inserting 118 vectors to relationships
INFO:lightrag:Writing graph with 126 nodes, 118 edges
你好
INFO:httpx:HTTP Request: POST http://0.0.0.0:8000/v1/chat/completions "HTTP/1.1 200 OK"
INFO:lightrag:Global query uses 59 entites, 60 relations, 3 text units
/usr/local/lib/python3.10/site-packages/lightrag/operate.py:1016: UserWarning: Low Level context is None. Return empty Low entity/relationship/source
  warnings.warn(
INFO:httpx:HTTP Request: POST http://0.0.0.0:8000/v1/chat/completions "HTTP/1.1 400 Bad Request"
An error occurred: Error code: 400 - {'object': 'error', 'message': 'could not broadcast input array from shape (535,) into shape (512,)', 'type': 'BadRequestError', 'param': None, 'code': 400}