TypeError: BaseTool.__call__() missing 1 required positional argument: 'tool_input'

[artificialai223]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain.pydantic_v1 import BaseModel, Field
from langchain.tools import BaseTool, StructuredTool, tool
import requests


def weather_lookup(location: str) -> str:
    """Find the weather of a location."""
    r = requests.get(
        'https://api.weatherapi.com/v1/current.json?q=' + location + '&key=redacted')
    return r


WeatherLookup = StructuredTool.from_function(
    func=weather_lookup,
    name="tool_weather_information",
    description="""A tool designed to gather accurate information on weather around the globe. 
    Useful for when you need to answer questions about the weather. 
    Input should be the location.""",
    handle_tool_error=True,
    # coroutine= ... <- you can specify an async method if desired as well
)

Description

I am trying to create a custom Weather tool and I get this error.
TypeError: BaseTool.call() missing 1 required positional argument: 'tool_input'

This is the code that adds the tool to the array.

# Add custom tools
weather_tool = WeatherLookup()

System Info

System Information

OS: Windows
OS Version: 10.0.22631
Python Version: 3.11.5 (tags/v3.11.5:cce6ba9, Aug 24 2023, 14:38:34) [MSC v.1936 64 bit (AMD64)]

Package Information

langchain_core: 0.1.40
langchain: 0.1.14
langchain_community: 0.0.31
langsmith: 0.1.40
langchain_google_genai: 1.0.1
langchain_mongodb: 0.1.3
langchain_openai: 0.1.1
langchain_text_splitters: 0.0.1
langserve: 0.0.51

Packages not installed (Not Necessarily a Problem)

The following packages were not found:

langgraph

[dosubot[bot]]

Hey @artificialai223, great to see you diving into the LangChain framework again! Hope you've been doing well since our last chat.

Based on the error message you provided, it seems like you're trying to instantiate your WeatherLookup tool without providing the required tool_input argument. The __call__ method in the BaseTool class, which your WeatherLookup tool is based on, requires this argument.

Here's the definition of the __call__ method in the BaseTool class:

def __call__(self, tool_input: str, callbacks: Callbacks = None) -> str:
    """Make tool callable."""
    return self.run(tool_input, callbacks=callbacks)

As you can see, this method expects a tool_input argument of type str and an optional callbacks argument. The error you're encountering suggests that you're not providing the tool_input argument when you're trying to use your custom Weather tool.

To resolve this issue, you should call your custom Weather tool with the required tool_input argument. For example:

# Assuming weather_tool is an instance of your custom Weather tool
weather_tool_result = weather_tool("your_tool_input_here")

Replace "your_tool_input_here" with the actual input your Weather tool expects. This should resolve the TypeError you're encountering.

You can find more information about the __call__ method in the BaseTool class in the LangChain codebase.

Please let me know if this resolves your issue or if you need further assistance.

Sources

• libs/core/langchain_core/tools.py

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[artificialai223]

@dosu-bot I am trying to make it so the AI can call it via OpenAI tools parser. Could you rewrite my tool file for this?

from langchain.pydantic_v1 import BaseModel, Field
from langchain.tools import BaseTool, StructuredTool, tool
import requests

def weather_lookup(location: str) -> str:
"""Find the weather of a location."""
r = requests.get(
'https://api.weatherapi.com/v1/current.json?q=' + location + '&key=ac8f6ec5fad649d88e1184551240804')
return r

WeatherLookup = StructuredTool.from_function(
func=weather_lookup,
name="tool_weather_information",
description="""A tool designed to gather accurate information on weather around the globe.
Useful for when you need to answer questions about the weather.
Input should be the location.""",
handle_tool_error=True,
# coroutine= ... <- you can specify an async method if desired as well
)

[dosubot[bot]]

Based on the code snippet you provided, it seems like you're trying to use the StructuredTool.from_function method to create a tool from your weather_lookup function. However, the TypeError you're encountering suggests that the BaseTool.__call__() method is missing a required positional argument: 'tool_input'.

The BaseTool.__call__() method in LangChain is designed to make the tool callable and it expects a tool_input argument. This tool_input is typically a string or a dictionary that contains the input data for the tool.

To fix the TypeError, you need to ensure that your weather_lookup function accepts a tool_input argument and uses it appropriately. Here's how you can modify your weather_lookup function:

def weather_lookup(tool_input: str) -> str:
    """Find the weather of a location."""
    r = requests.get(
        'https://api.weatherapi.com/v1/current.json?q=' + tool_input + '&key=ac8f6ec5fad649d88e1184551240804')
    return r.json()  # Assuming you want to return the JSON response

Then, you can create your WeatherLookup tool as follows:

WeatherLookup = StructuredTool.from_function(
    func=weather_lookup,
    name="tool_weather_information",
    description="""A tool designed to gather accurate information on weather around the globe. 
    Useful for when you need to answer questions about the weather. 
    Input should be the location.""",
    handle_tool_error=True,
    # coroutine= ... <- you can specify an async method if desired as well
)

Now, when you call WeatherLookup with a location string, it should work as expected:

weather_info = WeatherLookup('New York')
print(weather_info)

This should fix the TypeError you're encountering and ensure that your custom Weather tool is compatible with the OpenAI tools parser.

For more information on how to create and use tools in LangChain, you can refer to the LangChain Tools documentation and the source code for the BaseTool class.

I hope this helps! If you have any further questions or run into any issues, feel free to ask.

Sources

• libs/langchain/langchain/tools/openweathermap/tool.py
• libs/core/langchain_core/tools.py

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[liugddx]

Try this

import requests
from langchain.tools import StructuredTool


def weather_lookup(location: str) -> str:
    """Find the weather of a location."""
    r = requests.get(
        'https://api.weatherapi.com/v1/current.json?q=' + location + '&key=redacted')
    return r.json()


weather_tool = StructuredTool.from_function(
    func=weather_lookup,
    name="tool_weather_information",
    description="""A tool designed to gather accurate information on weather around the globe. 
    Useful for when you need to answer questions about the weather. 
    Input should be the location.""",
    handle_tool_error=True,
    # coroutine= ... <- you can specify an async method if desired as well
)

print(weather_tool.run("London"))

[artificialai223]

I resolved it, thanks for the help though. I copied the code from the Tavily tool and adapted it heavily.

…

On Tue, 9 Apr 2024, 06:30 Guangdong Liu, ***@***.***> wrote: Try this import requests from langchain.tools import StructuredTool def weather_lookup(location: str) -> str: """Find the weather of a location.""" r = requests.get( 'https://api.weatherapi.com/v1/current.json?q=' + location + '&key=redacted') return r.json() weather_tool = StructuredTool.from_function( func=weather_lookup, name="tool_weather_information", description="""A tool designed to gather accurate information on weather around the globe. Useful for when you need to answer questions about the weather. Input should be the location.""", handle_tool_error=True, # coroutine= ... <- you can specify an async method if desired as well ) print(weather_tool.run("London")) — Reply to this email directly, view it on GitHub <#20175 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ALK5WVEN4JDZZZWZQIHUQKTY4N4IZAVCNFSM6AAAAABF5HUVI6VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TANJUGI4DI> . You are receiving this because you were mentioned.Message ID: ***@***.*** com>

[Daviduche03]

hey can you link the code from the Tavily tool here please?

[artificialai223]

I can't since I used PyCharm to jump to source and I have heavily modified it to my needs now. You just have to find a sample tool and use that for future reference when making a tool.

…

On Sun, 14 Apr 2024, 01:06 David uche, ***@***.***> wrote: hey can you link the code from the Tavily tool here please? — Reply to this email directly, view it on GitHub <#20175 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ALK5WVD5E2LPEDCTSE66GQDY5HB7FAVCNFSM6AAAAABF5HUVI6VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TCMBWGY2TS> . You are receiving this because you were mentioned.Message ID: ***@***.*** com>

[keshavvarma]

Simply providing the input in the form of dict should work. weather_tool.run({"location" : "London"})
reference link: https://python.langchain.com/v0.2/docs/how_to/custom_tools/#structuredtool

[khteh]

    client = genai.Client(api_key=os.environ.get("GEMINI_API_KEY"))
    model_id = "gemini-2.0-flash"
    google_search_tool = Tool(
        google_search = GoogleSearch()
    )
    response = client.models.generate_content(
        model=model_id,
        contents=query,
        config=GenerateContentConfig(
            tools=[google_search_tool],
            response_modalities=["TEXT"],
        )
    )

Deprecation warning:

/usr/src/Python/rag-agent/src/rag_agent/Tools.py:51: LangChainDeprecationWarning: The method `BaseTool.__call__` was deprecated in langchain-core 0.1.47 and will be removed in 1.0. Use :meth:`~invoke` instead.

Error:

Error: TypeError("BaseTool.__call__() missing 1 required positional argument: 'tool_input'")