search

run-llama / llama_index

LlamaIndex is a data framework for your LLM applications
https://docs.llamaindex.ai
MIT License
35.36k stars 4.98k forks source link

在创建知识图谱的代码中，执行到retrieve时，发生了下面的错误，表示ValueError: too many values to unpack (expected 2) 通常表示在解包操作中，右侧提供的值数量多于左侧预期接收的变量数量 #15008

Open li-hhhh opened 1 month ago

li-hhhh commented 1 month ago

Question Validation

Question

我的代码如下： from llama_index.core import KnowledgeGraphIndex,SimpleDirectoryReader,Settings

from llama_index.graph_stores.neo4j import Neo4jGraphStore from llama_index.core.node_parser import SentenceSplitter from llama_index.core import PromptTemplate from llama_index.llms.huggingface import HuggingFaceLLM from llama_index.embeddings.huggingface import HuggingFaceEmbedding from llama_index.core import StorageContext, load_index_from_storage import os import torch import json import re #

llm = HuggingFaceLLM(

tokenizer_name='/home/cdhd/Desktop/glm4/glm-4-9b-chat',

model_name='/home/cdhd/Desktop/glm4/glm-4-9b-chat',

)

SYSTEM_PROMPT = """你是一个知识图谱查询专家，请将查询到的内容整理成完整的一句话返回，不添加其他的内容和自己的理解. """

query_wrapper_prompt = PromptTemplate( "[INST]<>\n" + SYSTEM_PROMPT + "<>\n\n{query_str}[/INST] " )

llm = HuggingFaceLLM( context_window=4096, max_new_tokens=1024, generate_kwargs={"temperature": 0.1, "do_sample": True}, query_wrapper_prompt=query_wrapper_prompt,

tokenizer_name='/home/cdhd/PythonCode/pythonProject/model/gpt2',

# model_name='/home/cdhd/PythonCode/pythonProject/model/gpt2',
tokenizer_name='/home/huidao/SDHD_AI_project/glm4chat/glm-4-9b-chat',
model_name='/home/huidao/SDHD_AI_project/glm4chat/glm-4-9b-chat',
device_map="auto",
# change these settings below depending on your GPU
model_kwargs={"torch_dtype": torch.float16},

)

embed= HuggingFaceEmbedding( model_name='/home/huidao/SDHD_AI_project/flask_project/KnowledgeBase/models/bge-large-zh-v1.5/models--BAAI--bge-large-zh-v1.5/snapshots/79e7739b6ab944e86d6171e44d24c997fc1e0116',

)

def graph_ground_process(documents,question,flag):

# 指定要检查的路径
path = './ground'
# 初始化Neo4j图存储
graph_store = Neo4jGraphStore(
    username="neo4j",
    password="12345678",
    url="bolt://localhost:7687",
    database="ground"
)

# 创建存储上下文
storage_context = StorageContext.from_defaults(graph_store=graph_store)

Settings.llm = llm

Settings.embed_model = embed

# node_parser = SentenceSplitter()
# nodes = node_parser.get_nodes_from_documents(documents)

if os.path.exists(path) and os.path.isdir(path) and (flag == True):
    print("exit!!!!!!")
    index = load_index_from_storage(
        StorageContext.from_defaults(persist_dir=path)
    )
    # embed_model=embed
    retriever = index.as_retriever(retriever_mode="hybrid",embed_model=embed)
    ans = retriever.retrieve(question)
    print("1",ans)
    for node in ans:
        text = node.metadata['kg_rel_texts']
        print(text)
    query_engine = index.as_query_engine(
        include_text=True, response_mode="tree_summarize",embedding_mode="hybrid",

    )
    response = query_engine.query(question)
    response_str = response.__str__()

    print("hh", response_str)

    result = re.split(r"[/INST]|<</INST>>", response_str)
    result = result[0]
    print("result:", result)

else:

    print("no exit!!!!!!!")
    # 清理数据集
    graph_store.query(
        """
    MATCH (n) DETACH DELETE n
    """
    )

    # 初始化一个空的索引
    index = KnowledgeGraphIndex.from_documents(
        documents,
        # [],
        storage_context=storage_context,
        max_triplets_per_chunk=None,
        include_embeddings=True,
    )

    json_filename = '/home/huidao/KnowledgeProject/pythonProject/GroundSystem/Data_Json/data.json'
    with open(json_filename, 'r' ,encoding='utf-8') as file:
        data = json.load(file)
        file.close()

    tups = [(item['first_entity'],item['relation'],item['two_entity']) for item in data]

    for tup in tups:
        # index.upsert_triplet_and_node(tup, nodes[0])
        index.upsert_triplet(tup,include_embeddings=True)

    # embed_model=embed_model
    retriever = index.as_retriever(retriever_mode="hybrid",embed_model=embed,similarity_top_k=10)
    print("1111")
    ans = retriever.retrieve(question)
    print("1",ans)
    for node in ans:
        text = node.metadata['kg_rel_texts']
        print(text)
    query_engine = index.as_query_engine(
        include_text=True, response_mode="tree_summarize",embedding_mode="hybrid",
        similarity_top_k = 10,
    )
    response = query_engine.query(question)

    response_str = response.__str__()

    print("hh",response_str )

    result = re.split(r"[/INST]|<</INST>>", response_str)
    result = result[0]
    print("result:",result)
    # display(Markdown(f"<b>{response}</b>"))
    # save and load
    index.storage_context.persist(persist_dir=path)

return result

检查是否直接运行了此脚本

if name == "main": documents = SimpleDirectoryReader( '/home/huidao/KnowledgeProject/pythonProject/地面流程' ).load_data() question = "节流阀的分类有哪几种，分别具有哪些特点？" graph_ground_process(documents,question,False)

报错信息：
Traceback (most recent call last):

File "/home/huidao/KnowledgeProject/pythonProject/GroundSystem/ground_demo.py", line 160, in graph_ground_process(documents,question,False) File "/home/huidao/KnowledgeProject/pythonProject/GroundSystem/ground_demo.py", line 105, in graph_ground_process index = KnowledgeGraphIndex.from_documents( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/indices/base.py", line 145, in from_documents return cls( ^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/indices/knowledge_graph/base.py", line 109, in init super().init( File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/indices/base.py", line 94, in init index_struct = self.build_index_from_nodes( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/indices/base.py", line 216, in build_index_from_nodes return self._build_index_from_nodes(nodes, build_kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/indices/knowledge_graph/base.py", line 215, in _build_index_from_nodes triplets = self._extract_triplets( ^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/indices/knowledge_graph/base.py", line 159, in _extract_triplets return self._llm_extract_triplets(text) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/indices/knowledge_graph/base.py", line 163, in _llm_extract_triplets response = self._llm.predict( ^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/instrumentation/dispatcher.py", line 230, in wrapper result = func(*args, *kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/llms/llm.py", line 438, in predict response = self.complete(formatted_prompt, formatted=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/instrumentation/dispatcher.py", line 230, in wrapper result = func(args, kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/core/llms/callbacks.py", line 429, in wrapped_llm_predict f_return_val = f(_self, *args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/llama_index/llms/huggingface/base.py", line 360, in complete tokens = self._model.generate( ^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context return func(*args, *kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/transformers/generation/utils.py", line 1914, in generate result = self._sample( ^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/transformers/generation/utils.py", line 2651, in _sample outputs = self( ^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl return self._call_impl(args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl return forward_call(*args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/.cache/huggingface/modules/transformers_modules/glm-4-9b-chat/modeling_chatglm.py", line 878, in forward transformer_outputs = self.transformer( ^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl return self._call_impl(*args, *kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl return forward_call(args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/.cache/huggingface/modules/transformers_modules/glm-4-9b-chat/modeling_chatglm.py", line 774, in forward hidden_states, presents, all_hidden_states, all_self_attentions = self.encoder( ^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl return self._call_impl(*args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl return forward_call(*args, *kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/.cache/huggingface/modules/transformers_modules/glm-4-9b-chat/modeling_chatglm.py", line 607, in forward layer_ret = layer( ^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl return self._call_impl(args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl return forward_call(*args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/.cache/huggingface/modules/transformers_modules/glm-4-9b-chat/modeling_chatglm.py", line 510, in forward attention_output, kv_cache = self.self_attention( ^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl return self._call_impl(*args, *kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/KnowledgeProject/pythonProject/.venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl return forward_call(args, kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/huidao/.cache/huggingface/modules/transformers_modules/glm-4-9b-chat/modeling_chatglm.py", line 376, in forward cache_k, cache_v = kv_cache ^^^^^^^^^^^^^^^^ ValueError: too many values to unpack (expected 2)

这个具体应该如何修改代码解决我的问题。

logan-markewich commented 1 month ago

Seems like an issue with huggingface/transformers, not llamaindex

li-hhhh commented 1 month ago

Seems like an issue with huggingface/transformers, not llamaindex

好的，谢谢。真的很奇怪，我之前相同的代码都能正常运行，今天突然发生这样的问题。