add script for comparing automl eval

[Samoed]

No description provided.

[Copilot]

Pull Request Overview

This PR adds a new script for comparing AutoML evaluation frameworks by introducing various evaluation functions for frameworks such as Fedot, H2O, LAMA, GAMA, and GlueOn. The script accepts command-line arguments for dataset and framework, loads data from Hugging Face’s datasets, performs model evaluation using the specified framework, and logs results via wandb.

Comments suppressed due to low confidence (1)

scripts/automl_eval.py:44

• The function name 'evalute_fedot' appears to be misspelled. Consider renaming it to 'evaluate_fedot' for clarity and consistency.

def evalute_fedot(train_df: pd.DataFrame, test_df: pd.DataFrame):

[Copilot]

The docstring for load_data specifies a return type of 'DatasetDict' despite the function actually returning two DataFrames. Update the docstring to accurately describe the returned values.

Suggested change

	Load a dataset from the Hugging Face datasets library.
	Args:
	dataset_name (str): The name of the dataset to load.
	Returns:
	DatasetDict: A dictionary containing the train, validation, and test splits of the dataset.
	Load a dataset from the Hugging Face datasets library and return it as pandas DataFrames.
	Args:
	dataset_name (str): The name of the dataset to load.
	Returns:
	tuple[pd.DataFrame, pd.DataFrame]: A tuple containing two DataFrames:
	- The first DataFrame contains the training data.
	- The second DataFrame contains the testing data.