导入OpenAI与LangChain后GPT聊天机器人报错，求解决方法

• 目标：基于OpenAI API开发一款可读取PDF、TXT等文件，并能基于文件内容回答问题的聊天机器人
• 开发依据：参考相关教程实现，使用代码如下：

from gpt_index import SimpleDirectoryReader, GPTListIndex, GPTSimpleVectorIndex, LLMPredictor, PromptHelper, ServiceContext
from langchain.chat_models import ChatOpenAI
import gradio as gr
import sys
import os

os.environ["OPENAI_API_KEY"] = 'yourapikey'

def construct_index(directory_path):
    max_input_size = 4096
    num_outputs = 512
    max_chunk_overlap = 20
    chunk_size_limit = 600

    prompt_helper = PromptHelper(max_input_size, num_outputs, max_chunk_overlap, chunk_size_limit=chunk_size_limit)

    llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0.7, model_name="gpt-3.5-turbo", max_tokens=num_outputs))

    documents = SimpleDirectoryReader(directory_path).load_data()

    index = GPTSimpleVectorIndex(documents, llm_predictor=llm_predictor, prompt_helper=prompt_helper)

    index.save_to_disk('index.json')

    return index

def chatbot(input_text):
    index = GPTSimpleVectorIndex.load_from_disk('index.json')
    response = index.query(input_text, response_mode="compact")
    return response.response

iface = gr.Interface(fn=chatbot,
                     inputs=gr.components.Textbox(lines=7, label="Enter your text"),
                     outputs="text",
                     title="Custom-trained AI Chatbot")

index = construct_index("docs")
iface.launch(share=True)

运行代码后触发如下错误：

File "C:\Users\USER\desktop\cht\app.py", line 1, in <module>
    from gpt_index import SimpleDirectoryReader, GPTListIndex, GPTSimpleVectorIndex, LLMPredictor, PromptHelper, ServiceContext
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\__init__.py", line 18, in <module>
    from gpt_index.indices.common.struct_store.base import SQLDocumentContextBuilder
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\indices\__init__.py", line 4, in <module>
    from gpt_index.indices.keyword_table.base import GPTKeywordTableIndex
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\indices\keyword_table\__init__.py", line 4, in <module>
    from gpt_index.indices.keyword_table.base import GPTKeywordTableIndex
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\indices\keyword_table\base.py", line 16, in <module>
    from gpt_index.indices.base import DOCUMENTS_INPUT, BaseGPTIndex
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\indices\base.py", line 23, in <module>
    from gpt_index.indices.prompt_helper import PromptHelper
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\indices\prompt_helper.py", line 12, in <module>
    from gpt_index.langchain_helpers.chain_wrapper import LLMPredictor
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\langchain_helpers\chain_wrapper.py", line 13, in <module>
    from gpt_index.prompts.base import Prompt
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\prompts\__init__.py", line 3, in <module>
    from gpt_index.prompts.base import Prompt
  File "C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\gpt_index\prompts\base.py", line 9, in <module>
    from langchain.schema import BaseLanguageModel
ImportError: cannot import name 'BaseLanguageModel' from 'langchain.schema' (C:\Users\USER\AppData\Local\Programs\Python\Python311\Lib\site-packages\langchain\schema.py)

该错误由langchain与gpt-index（现更名为llama-index）版本不兼容导致：旧版gpt-index依赖langchain中位于langchain.schema的BaseLanguageModel类，但新版langchain已将该类迁移至langchain.base_language模块，旧版gpt-index无法适配此变更。

提供两种可行解决方案：

方案一：降级langchain至兼容版本 #

安装与旧版gpt-index匹配的langchain版本（如0.0.240，此版本仍保留BaseLanguageModel在langchain.schema中）：

pip install langchain==0.0.240

执行完成后重新运行原代码即可。

方案二：升级gpt-index至最新版（推荐） #

gpt-index现已更名为llama-index，新版已适配langchain的API变更，操作步骤如下：

1. 安装最新版llama-index：

pip install llama-index

2. 修改代码以适配新版API：

from llama_index import SimpleDirectoryReader, VectorStoreIndex, LLMPredictor, PromptHelper, ServiceContext
from langchain.chat_models import ChatOpenAI
import gradio as gr
import sys
import os

os.environ["OPENAI_API_KEY"] = 'yourapikey'

def construct_index(directory_path):
    max_input_size = 4096
    num_outputs = 512
    max_chunk_overlap = 20
    chunk_size_limit = 600

    prompt_helper = PromptHelper(max_input_size, num_outputs, max_chunk_overlap, chunk_size_limit=chunk_size_limit)

    llm_predictor = LLMPredictor(llm=ChatOpenAI(temperature=0.7, model_name="gpt-3.5-turbo", max_tokens=num_outputs))

    documents = SimpleDirectoryReader(directory_path).load_data()

    # 替换旧版GPTSimpleVectorIndex为新版VectorStoreIndex
    index = VectorStoreIndex.from_documents(documents, llm_predictor=llm_predictor, prompt_helper=prompt_helper)

    # 新版索引保存方式
    index.storage_context.persist(persist_dir="./")

    return index

def chatbot(input_text):
    # 新版索引加载方式
    from llama_index import StorageContext, load_index_from_storage
    storage_context = StorageContext.from_defaults(persist_dir="./")
    index = load_index_from_storage(storage_context)
    
    response = index.query(input_text, response_mode="compact")
    return response.response

iface = gr.Interface(fn=chatbot,
                     inputs=gr.components.Textbox(lines=7, label="Enter your text"),
                     outputs="text",
                     title="Custom-trained AI Chatbot")

index = construct_index("docs")
iface.launch(share=True)

3. 运行修改后的代码即可。

内容的提问来源于stack exchange，提问作者Mohammad saabir Mustaaan