[Python] Why is this happening, even though I’ve already defined a Pydantic input model using...

I am using LangChain’s Tool.from_function() to define a tool called derivative, and I’ve also provided a Pydantic BaseModel (Input_structure) using the args_schema parameter:

class Input_structure(BaseModel):
expression: str
variable: str

def compute_derivative(expression: str, variable: str) -> str:
...

derivative_tool = Tool.from_function(
func=compute_derivative,
name="derivative",
description="Compute derivative",
args_schema=Input_structure
)


However, when I chain this tool like so:

result = derivative_tool.invoke(('x^2 + 10', 'x'))


I get the following error:

ToolException: Too many arguments to single-input tool derivative.
Consider using StructuredTool instead. Args: ['x^2 + 10', 'x']

Continue reading...