Azure Code Interpreter 工具规范¶

此示例介绍了如何配置和使用 Azure Code Interpreter 工具规范（由 Azure Dynamic Sessions 提供支持）。

在 [ ]

%pip install llama-index
%pip install llama-index-llms-azure
%pip install llama-index-tools-azure-code-interpreter

%pip install llama-index %pip install llama-index-llms-azure %pip install llama-index-tools-azure-code-interpreter

在 [ ]

# Setup Azure OpenAI Agent
from llama_index.llms.azure_openai import AzureOpenAI

api_key = "your-azure-openai-api-key"
azure_endpoint = "your-azure-openai-endpoint"
api_version = "azure-api-version"

# 设置 Azure OpenAI Agent from llama_index.llms.azure_openai import AzureOpenAI api_key = "your-azure-openai-api-key" azure_endpoint = "your-azure-openai-endpoint" api_version = "azure-api-version"

在 [ ]

# Import the AzureCodeInterpreterToolSpec from llama_index
from llama_index.tools.azure_code_interpreter import (
    AzureCodeInterpreterToolSpec,
)

# Import the ReActAgent
from llama_index.core.agent import ReActAgent

# Create the AzureCodeInterpreterToolSpec with the pool_management_endpoint set to your session management endpoint
# It is optional to set the local_save_path, but it is recommended to set it to a path where the tool can automatically save any intermediate data generated from Python code's output.
azure_code_interpreter_spec = AzureCodeInterpreterToolSpec(
    pool_management_endpoint="your-pool-management-endpoint",
    local_save_path="local-file-path-to-save-intermediate-data",
)

llm = AzureOpenAI(
    model="gpt-35-turbo",
    deployment_name="gpt-35-deploy",
    api_key=api_key,
    azure_endpoint=azure_endpoint,
    api_version=api_version,
)

# Create the ReActAgent and inject the tools defined in the AzureDynamicSessionsToolSpec
agent = ReActAgent.from_tools(
    azure_code_interpreter_spec.to_tool_list(), llm=llm, verbose=True
)

# 从 llama_index 导入 AzureCodeInterpreterToolSpec from llama_index.tools.azure_code_interpreter import ( AzureCodeInterpreterToolSpec, ) # 从 llama_index.core.agent 导入 ReActAgent from llama_index.core.agent import ReActAgent # 创建 AzureCodeInterpreterToolSpec 并将 pool_management_endpoint 设置为您的会话管理 endpoint # 设置 local_save_path 是可选的，但建议将其设置为一个路径，以便工具可以自动保存 Python 代码输出产生的任何中间数据。 azure_code_interpreter_spec = AzureCodeInterpreterToolSpec( pool_management_endpoint="your-pool-management-endpoint", local_save_path="local-file-path-to-save-intermediate-data", ) llm = AzureOpenAI( model="gpt-35-turbo", deployment_name="gpt-35-deploy", api_key=api_key, azure_endpoint=azure_endpoint, api_version=api_version, ) # 创建 ReActAgent 并注入 AzureDynamicSessionsToolSpec 中定义的工具 agent = ReActAgent.from_tools( azure_code_interpreter_spec.to_tool_list(), llm=llm, verbose=True )

在 [ ]

# You can use the code interpreter directly without the LLM agent.
print(azure_code_interpreter_spec.code_interpreter("1+1"))

# 您可以不使用 LLM agent 直接使用代码解释器。 print(azure_code_interpreter_spec.code_interpreter("1+1"))

{'$id': '1', 'status': 'Success', 'stdout': '', 'stderr': '', 'result': 2, 'executionTimeInMilliseconds': 11}

在 [ ]

# Test the agent with simple answers that could leverage Python codes
print(agent.chat("Tell me the current time in Seattle."))

# 使用可以通过 Python 代码利用的简单答案测试 agent print(agent.chat("告诉我西雅图的当前时间。"))

Thought: To provide the current time in Seattle, I need to calculate it based on the current UTC time and adjust for Seattle's time zone, which is Pacific Daylight Time (PDT) during daylight saving time and Pacific Standard Time (PST) outside of daylight saving time. PDT is UTC-7, and PST is UTC-8. I can use the code interpreter tool to get the current UTC time and adjust it accordingly.
Action: code_interpreter
Action Input: {'python_code': "from datetime import datetime, timedelta; import pytz; utc_now = datetime.now(pytz.utc); seattle_time = utc_now.astimezone(pytz.timezone('America/Los_Angeles')); seattle_time.strftime('%Y-%m-%d %H:%M:%S %Z%z')"}
Observation: {'$id': '1', 'status': 'Success', 'stdout': '', 'stderr': '', 'result': '2024-05-04 13:54:09 PDT-0700', 'executionTimeInMilliseconds': 120}
Thought: I can answer without using any more tools. I'll use the user's language to answer.
Answer: The current time in Seattle is 2024-05-04 13:54:09 PDT.
The current time in Seattle is 2024-05-04 13:54:09 PDT.

在 [ ]

# Upload a sample temperature file of a day in Redmond Washington and ask a question about it
res = azure_code_interpreter_spec.upload_file(
    local_file_path="./TemperatureData.csv"
)
if len(res) != 0:
    print(
        agent.chat("Find the highest temperature in the file that I uploaded.")
    )

# 上传华盛顿州雷德蒙德某天的温度示例文件并提问 res = azure_code_interpreter_spec.upload_file( local_file_path="./TemperatureData.csv" ) if len(res) != 0: print( agent.chat("找到我上传的文件中的最高温度。") )

Thought: I need to use the list_files tool to get the metadata for the uploaded file, and then use python to read the file and find the highest temperature.
Action: list_files
Action Input: {}
Observation: [RemoteFileMetadata(filename='TemperatureData.csv', size_in_bytes=514, file_full_path='/mnt/data/TemperatureData.csv')]
Thought: I have the metadata for the file. I need to use python to read the file and find the highest temperature.
Action: code_interpreter
Action Input: {'python_code': "import csv\n\nwith open('/mnt/data/TemperatureData.csv', 'r') as f:\n    reader = csv.reader(f)\n    next(reader)\n    highest_temp = float('-inf')\n    for row in reader:\n        temp = float(row[1])\n        if temp > highest_temp:\n            highest_temp = temp\nprint(highest_temp)"}
Observation: {'$id': '1', 'status': 'Success', 'stdout': '12.4\n', 'stderr': '', 'result': '', 'executionTimeInMilliseconds': 26}
Thought: I have the highest temperature. I can answer the question.
Answer: The highest temperature in the file is 12.4 degrees.
The highest temperature in the file is 12.4 degrees.

在 [ ]

# Ask the LLM to draw a diagram based on the uploaded file.
# Because the local_save_path is set, the diagram data will be automatically saved to the local_save_path.
print(
    agent.chat(
        "Use the temperature data that I uploaded, create a temperature curve."
    )
)

# 请求 LLM 根据上传的文件绘制图表。 # 因为设置了 local_save_path，图表数据将自动保存到 local_save_path。 print( agent.chat( "使用我上传的温度数据，创建温度曲线。" ) )

Thought: To proceed, I need to list the files in the session pool to identify the uploaded file containing the temperature data.
Action: list_files
Action Input: {}
Observation: [RemoteFileMetadata(filename='TemperatureData.csv', size_in_bytes=514, file_full_path='/mnt/data/TemperatureData.csv')]
Thought: The file containing the temperature data is identified as 'TemperatureData.csv'. I need to read this file to extract the temperature data for plotting the temperature curve.
Action: code_interpreter
Action Input: {'python_code': "import pandas as pd\n\ndf = pd.read_csv('/mnt/data/TemperatureData.csv')\nprint(df.head())"}
Observation: {'$id': '1', 'status': 'Success', 'stdout': '             Time  Temperature\n0  4/11/2024 0:00          6.5\n1  4/11/2024 1:00          5.2\n2  4/11/2024 2:00          4.8\n3  4/11/2024 3:00          5.1\n4  4/11/2024 4:00          5.7\n', 'stderr': '', 'result': '', 'executionTimeInMilliseconds': 2103}
Thought: The data has been successfully loaded and previewed, showing columns for Time and Temperature. I will now proceed to plot the temperature curve using this data.
Action: code_interpreter
Action Input: {'python_code': "import pandas as pd\nimport matplotlib.pyplot as plt\n\ndf = pd.read_csv('/mnt/data/TemperatureData.csv')\n\nplt.figure(figsize=(10, 6))\nplt.plot(pd.to_datetime(df['Time']), df['Temperature'], marker='o', linestyle='-', color='b')\nplt.title('Temperature Curve')\nplt.xlabel('Time')\nplt.ylabel('Temperature (°C)')\nplt.xticks(rotation=45)\nplt.tight_layout()\nplt.savefig('/mnt/data/TemperatureCurve.png')\nprint('/mnt/data/TemperatureCurve.png')"}
Observation: {'$id': '1', 'status': 'Success', 'stdout': '/mnt/data/TemperatureCurve.png\n', 'stderr': '', 'result': {'type': 'image', 'format': 'png', 'saved_to_local_path': True}, 'executionTimeInMilliseconds': 3810}
Thought: I can answer without using any more tools. I'll use the user's language to answer.
Answer: The temperature curve has been successfully created and saved. You can view and download the image from [here](sandbox:/mnt/data/TemperatureCurve.png).
The temperature curve has been successfully created and saved. You can view and download the image from [here](sandbox:/mnt/data/TemperatureCurve.png).

在 [ ]

# Ask the LLM to perform modifications on the sample temperature file
print(
    agent.chat(
        "Rearrange the temperature data in a descending order and save it back to the original csv file."
    )
)

# 请求 LLM 对示例温度文件进行修改 print( agent.chat( "将温度数据按降序重新排列并保存回原始 csv 文件。" ) )

Thought: To proceed, I need to first list the files in the session pool to identify the uploaded temperature data file.
Action: list_files
Action Input: {}
Observation: [RemoteFileMetadata(filename='TemperatureCurve.png', size_in_bytes=35512, file_full_path='/mnt/data/TemperatureCurve.png'), RemoteFileMetadata(filename='TemperatureData.csv', size_in_bytes=514, file_full_path='/mnt/data/TemperatureData.csv')]
Thought: The temperature data file is identified as 'TemperatureData.csv'. I need to read its contents, rearrange the temperature data in descending order, and then save the changes back to the original CSV file.
Action: code_interpreter
Action Input: {'python_code': "import pandas as pd\n\ndf = pd.read_csv('/mnt/data/TemperatureData.csv')\ndf_sorted = df.sort_values(by='Temperature', ascending=False)\ndf_sorted.to_csv('/mnt/data/TemperatureData.csv', index=False)\n'filename': 'TemperatureData.csv'"}
Observation: {'$id': '1', 'status': 'Failure', 'stdout': '', 'stderr': "illegal target for annotation (3407605186.py, line 6)  Cell In[2363], line 6\n    'filename': 'TemperatureData.csv'\n    ^\nSyntaxError: illegal target for annotation\n\n", 'result': None, 'executionTimeInMilliseconds': 9}
Thought: There was a syntax error in the Python code due to an incorrect line. I need to correct the code and execute it again to sort the temperature data and save it back to the CSV file.
Action: code_interpreter
Action Input: {'python_code': "import pandas as pd\n\ndf = pd.read_csv('/mnt/data/TemperatureData.csv')\ndf_sorted = df.sort_values(by='Temperature', ascending=False)\ndf_sorted.to_csv('/mnt/data/TemperatureData.csv', index=False)"}
Observation: {'$id': '1', 'status': 'Success', 'stdout': '', 'stderr': '', 'result': '', 'executionTimeInMilliseconds': 91}
Thought: I can answer without using any more tools. I'll use the user's language to answer.
Answer: The temperature data has been successfully rearranged in descending order and saved back to the original CSV file, 'TemperatureData.csv'.
The temperature data has been successfully rearranged in descending order and saved back to the original CSV file, 'TemperatureData.csv'.

在 [ ]

# Download the modified file
azure_code_interpreter_spec.download_file_to_local(
    remote_file_path="TemperatureData.csv",
    local_file_path="/.../SortedTemperatureData.csv",
)

# 下载修改后的文件 azure_code_interpreter_spec.download_file_to_local( remote_file_path="TemperatureData.csv", local_file_path="/.../SortedTemperatureData.csv", )

在 [ ]

# For comparison, print the first 10 lines of the original file
with open("/.../TemperatureData.csv", "r") as f:
    for i in range(10):
        print(f.readline().strip())

# 为了比较，打印原始文件的前 10 行 with open("/.../TemperatureData.csv", "r") as f: for i in range(10): print(f.readline().strip())

Time,Temperature
4/11/2024 0:00,6.5
4/11/2024 1:00,5.2
4/11/2024 2:00,4.8
4/11/2024 3:00,5.1
4/11/2024 4:00,5.7
4/11/2024 5:00,5.1
4/11/2024 6:00,4.5
4/11/2024 7:00,5.5
4/11/2024 8:00,5.3

在 [ ]

# For comparison, print the first 10 lines of the sorted file downloaded from session pool
with open("/.../SortedTemperatureData.csv", "r") as f:
    for i in range(10):
        print(f.readline().strip())

# 为了比较，打印从会话池下载的排序文件的前 10 行 with open("/.../SortedTemperatureData.csv", "r") as f: for i in range(10): print(f.readline().strip())

Time,Temperature
4/11/2024 20:00,12.4
4/11/2024 19:00,12.3
4/11/2024 17:00,12.3
4/11/2024 18:00,12.1
4/11/2024 16:00,11.7
4/11/2024 15:00,11.3
4/11/2024 21:00,10.9
4/11/2024 22:00,10.0
4/11/2024 23:00,9.4