【GPT入门】第31课 两行代码搞定模型debug调试

1. 两行代码搞定调试

from langchain.globals import set_debug,
set_debug(True)

from langchain_ollama import OllamaLLM
from langchain.globals import set_debug, set_verbose

set_debug(True)

#set_verbose(True)
llm = OllamaLLM(model="gpt-4")
print(llm.invoke("你是谁？"))

输出结果,就会带有详细信息 ：

C:\ProgramData\anaconda3\envs\gptLearning\python.exe "E:\workspace\gptLearning\gptLearning\ls10\06 ollama\03.debug.py" 
[llm/start] [llm:OllamaLLM] Entering LLM run with input:
{
  "prompts": [
    "你是谁？"
  ]
}
[llm/end] [llm:OllamaLLM] [10.98s] Exiting LLM run with output:
{
  "generations": [
    [
      {
        "text": "我是来自阿里云的大规模语言模型，我被命名为通义千问。我可以回答各种问题，包括但不限于科技、文化、生活、历史、地理等各个领域的问题。我还能够进行文本摘要、代码解析、图像生成、视频转码等各种任务。如果您有任何具体问题或需要某个特定任务的帮助，请随时告诉我，我会尽力为您提供最准确和最有用的答案和帮助。",
        "generation_info": {
          "model": "gpt-4",
          "created_at": "2025-04-02T21:53:46.7502575Z",
          "done": true,
          "done_reason": "stop",
          "total_duration": 10975642300,
          "load_duration": 3865007300,
          "prompt_eval_count": 11,
          "prompt_eval_duration": 1252263100,
          "eval_count": 83,
          "eval_duration": 5857248900,
          "response": "",
          "context": [
            151644,
            872,
            198,
            105043,
            100165,
            11319,
            151645,
            198,
            151644,
            77091,
            198,
            104198,
            101919,
            102661,
            99718,
            104197,
            100176,
            102064,
            104949,
            3837,
            35946,
            99250,
            116816,
            31935,
            64559,
            99320,
            56007,
            1773,
            109944,
            102104,
            100646,
            86119,
            3837,
            100630,
            116509,
            99602,
            5373,
            99348,
            5373,
            99424,
            5373,
            100022,
            5373,
            102752,
            49567,
            101284,
            100650,
            103936,
            1773,
            107228,
            100006,
            71817,
            108704,
            106208,
            5373,
            46100,
            106637,
            5373,
            107553,
            43959,
            5373,
            87140,
            46670,
            16476,
            113464,
            88802,
            1773,
            106870,
            110117,
            100398,
            86119,
            57191,
            85106,
            106168,
            105149,
            88802,
            116211,
            3837,
            14880,
            102422,
            106525,
            3837,
            105351,
            110121,
            113445,
            31235,
            102188,
            33108,
            108543,
            11622,
            111230,
            33108,
            100364,
            1773
          ]
        },
        "type": "Generation"
      }
    ]
  ],
  "llm_output": null,
  "run": null,
  "type": "LLMResult"
}
我是来自阿里云的大规模语言模型，我被命名为通义千问。我可以回答各种问题，包括但不限于科技、文化、生活、历史、地理等各个领域的问题。我还能够进行文本摘要、代码解析、图像生成、视频转码等各种任务。如果您有任何具体问题或需要某个特定任务的帮助，请随时告诉我，我会尽力为您提供最准确和最有用的答案和帮助。

Process finished with exit code 0