Initialize the Docker-based Jupyter Kernel Gateway executor.

Initialize experiment processor Args: model_kwargs: Model configuration parameters **kwargs: Additio

Initialize the agent experiment processor with a rich console if available

Initialize the reasoning experiment processor with a rich console if available

Implements len() function support

Representation for debugging

String representation

Displays correctly this type in an ipython notebook (ipython, colab, jupyter, ...)

Displays correctly this type in an ipython notebook (ipython, colab, jupyter, ...)

A decorator that adds a description to a function.

Add 'self' as first argument to a function definition if not present.

Recursively builds the agent tree.

Clean the units in the number.

Clean the units in the number.

Closes any visible modal or pop-up on the page. Use this to dismiss pop-up windows! This does not work on cookie consent banners.

Ensure cleanup on deletion.

Evaluate if the predicted answer matches the gold answer using the model Returns dict with score and explanation

Evaluate if the predicted answer matches the gold answer using the model Returns dict with score and explanation

Parse action from the LLM output Args: model_output (`str`): Output of the LLM split_token (`str`): Separato

Groups answers based on their canonical forms and finds the group with the largest sum of scores. Args: answers (list of str): A lis

Groups answers based on their canonical forms and finds the group with the largest number of elements. In case of a tie, returns the first oc

Creates a [`Tool`] from a gradio tool.