.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "build_tutorial/monitor.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        :ref:`Go to the end <sphx_glr_download_build_tutorial_monitor.py>`
        to download the full example code.

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_build_tutorial_monitor.py:


.. _configuring_and_monitoring:

配置和监控
==================================

AgentScope 的主入口是 `agentscope.init`，在这里您可以配置应用程序。

.. GENERATED FROM PYTHON SOURCE LINES 10-34

.. code-block:: Python


    import agentscope


    agentscope.init(
        model_configs=[  # 模型配置
            {
                "config_name": "my-qwen-max",
                "model_type": "dashscope_chat",
                "model_name": "qwen-max",
            },
        ],
        project="项目 Alpha",  # 项目名称
        name="测试-1",  # 运行时名称
        disable_saving=False,  # 是否禁用文件保存，推荐开启
        save_dir="./runs",  # 保存目录
        save_log=True,  # 是否保存日志
        save_code=False,  # 是否保存此次运行的代码
        save_api_invoke=False,  # 保存 API 调用
        cache_dir="~/.cache",  # 缓存目录，用于缓存 Embedding 和其它
        use_monitor=True,  # 是否监控 token 使用情况
        logger_level="INFO",  # 日志级别
    )





.. rst-class:: sphx-glr-script-out

 .. code-block:: none


    []



.. GENERATED FROM PYTHON SOURCE LINES 35-39

导出配置
--------------------------------
`state_dict` 方法可用于导出正在运行的应用程序的配置。


.. GENERATED FROM PYTHON SOURCE LINES 39-44

.. code-block:: Python


    import json

    print(json.dumps(agentscope.state_dict(), indent=2))





.. rst-class:: sphx-glr-script-out

 .. code-block:: none

    {
      "project": "\u9879\u76ee Alpha",
      "name": "\u6d4b\u8bd5-1",
      "disable_saving": false,
      "run_id": "run_20250310-033918_\u6d4b\u8bd5-1",
      "pid": 2021,
      "timestamp": "2025-03-10 03:39:18",
      "agentscope_version": "0.1.2",
      "file": {
        "save_log": true,
        "save_code": false,
        "save_api_invoke": false,
        "base_dir": null,
        "run_dir": "/home/runner/work/agentscope/agentscope/docs/tutorial/zh_CN/source/tutorial/runs/run_20250310-033918_\u6d4b\u8bd5-1",
        "cache_dir": "~/.cache"
      },
      "model": {
        "model_configs": {
          "my-qwen-vl": {
            "config_name": "my-qwen-vl",
            "model_type": "dashscope_multimodal",
            "model_name": "qwen-vl-max"
          },
          "qwen_config": {
            "model_type": "dashscope_chat",
            "config_name": "qwen_config",
            "model_name": "qwen-max"
          },
          "my-qwen-max": {
            "config_name": "my-qwen-max",
            "model_type": "dashscope_chat",
            "model_name": "qwen-max"
          }
        }
      },
      "logger": {
        "level": "INFO"
      },
      "studio": {
        "active": false,
        "studio_url": null
      },
      "monitor": {
        "use_monitor": true,
        "path_db": "/home/runner/work/agentscope/agentscope/docs/tutorial/zh_CN/source/tutorial/runs/run_20250310-033918_\u6d4b\u8bd5-1/agentscope.db"
      }
    }




.. GENERATED FROM PYTHON SOURCE LINES 45-50

运行监控
--------------------------
AgentScope 提供了 AgentScope Studio，这是一个 Web 可视化界面，用于监控和管理正在运行的应用程序和历史记录。
有关更多详细信息，请参阅 :ref:`visual` 部分。


.. GENERATED FROM PYTHON SOURCE LINES 52-58

.. _token_usage:

监控 Token 使用情况
------------------------
`print_llm_usage` 将打印并返回当前运行应用程序的 token 使用情况。


.. GENERATED FROM PYTHON SOURCE LINES 58-77

.. code-block:: Python


    from agentscope.models import DashScopeChatWrapper

    qwen_max = DashScopeChatWrapper(
        config_name="-",
        model_name="qwen-max",
    )
    qwen_plus = DashScopeChatWrapper(
        config_name="-",
        model_name="qwen-plus",
    )

    # 调用 qwen-max 和 qwen-plus 来模拟 token 使用情况
    _ = qwen_max([{"role": "user", "content": "Hi!"}])
    _ = qwen_plus([{"role": "user", "content": "Who are you?"}])

    usage = agentscope.print_llm_usage()

    print(json.dumps(usage, indent=2))




.. rst-class:: sphx-glr-script-out

 .. code-block:: none

    {
      "text_and_embedding": [
        {
          "model_name": "qwen-max",
          "times": 1,
          "prompt_tokens": 10,
          "completion_tokens": 9,
          "total_tokens": 19
        },
        {
          "model_name": "qwen-plus",
          "times": 1,
          "prompt_tokens": 12,
          "completion_tokens": 54,
          "total_tokens": 66
        }
      ],
      "image": []
    }





.. rst-class:: sphx-glr-timing

   **Total running time of the script:** (0 minutes 6.579 seconds)


.. _sphx_glr_download_build_tutorial_monitor.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: monitor.ipynb <monitor.ipynb>`

    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: monitor.py <monitor.py>`

    .. container:: sphx-glr-download sphx-glr-download-zip

      :download:`Download zipped: monitor.zip <monitor.zip>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_