{"kind":"task","effective_mode":"full","benchmark":{"kind":"benchmark","effective_mode":"full","slug":"bigcodebench","formal_name":"BigCodeBench","introduction":"BigCodeBench poses 1,140 function-level tasks drawn across 139 libraries. It tests whether a model can compose several real APIs correctly rather than write one self-contained function.","introduction_ja":"","introduction_en":"","category":"Category not supplied","task_count":null,"acquisition_status":"Acquisition status not supplied","official_url":"https://bigcode-bench.github.io/","indexing_mode":"index","profile":{"resources":[],"task_format":"","scoring":"","metric":"","size":"","answer_access":"","license":"","citation":"","maintainer":"","released":"","why_hard":"","related":[],"history":"","construction":"","limitations":"","saturation":"","scores":[]}},"task_id":"6217796a-ba40-59b1-beb2-0799a2380c67","task_key":"default--v0~2e1~2e0~5fhf--6217796a-ba40-59b1-beb2-0799a2380c67","task_revision_id":"1","upstream_id":"","short_description":"BigCodeBench v0.1.0_hf 6217796a-ba40-59b1-beb2-0799a2380c67","config":"default","split":"v0.1.0_hf","body":"{\"code_prompt\":\"import heapq\\nfrom scipy import stats\\ndef task_func(df, col1, col2, N=10):\\n\",\"complete_prompt\":\"import heapq\\nfrom scipy import stats\\n\\ndef task_func(df, col1, col2, N=10):\\n    \\\"\\\"\\\"\\n    Find the N largest absolute differences between the corresponding elements\\n    of two specified columns in a DataFrame, perform a t-Test on the elements\\n    with these differences, and return the calculated p-value.\\n\\n    Parameters:\\n    df (pandas.DataFrame): A DataFrame containing at least two numerical columns to compare.\\n    col1, col2 (str): Names of the columns to compare.\\n    N (int, optional): The number of largest differences to consider for the t-Test. Defaults to 10.\\n\\n    Returns:\\n    float: The p-value resulting from the t-Test on the elements with the N largest differences.\\n\\n    Raises:\\n    ValueError: If specified columns are not in the provided DataFrame.\\n    ValueError: If N is <= 1.\\n\\n    Requirements:\\n    - scipy.stats\\n    - heapq\\n\\n    Example:\\n    >>> df = pd.DataFrame({\\n    ...     'col1': [99, 86, 90, 70, 86, 95, 56, 98, 80, 81],\\n    ...     'col2': [21, 11, 21, 1, 26, 40, 4, 50, 34, 37]\\n    ... })\\n    >>> p_value = task_func(df, 'col1', 'col2', N=5)\\n    >>> print(p_value)    \\n    4.676251508205865e-06\\n\\n    >>> df = pd.DataFrame({\\n    ...    'col1': [1, 3, 4, 70],\\n    ...    'col2': [2, 3, 5, 1]\\n    ...     })\\n    >>> p_value = task_func(df, 'col1', 'col2', N=5)\\n    >>> print(p_value)\\n    0.3590111759771484\\n\\n\\n    \\\"\\\"\\\"\\n\",\"entry_point\":\"task_func\",\"instruct_prompt\":\"Find the N largest absolute differences between the corresponding elements of two specified columns in a DataFrame, perform a t-Test on the elements with these differences, and return the calculated p-value. >>> df = pd.DataFrame({ ...    'col1': [1, 3, 4, 70], ...    'col2': [2, 3, 5, 1] ...     }) >>> p_value = task_func(df, 'col1', 'col2', N=5) >>> print(p_value) 0.3590111759771484\\nThe function should raise the exception for: ValueError: If specified columns are not in the provided DataFrame. ValueError: If N is <= 1.\\nThe function should output with:\\n    float: The p-value resulting from the t-Test on the elements with the N largest differences.\\nYou should write self-contained code starting with:\\n```\\nimport heapq\\nfrom scipy import stats\\ndef task_func(df, col1, col2, N=10):\\n```\",\"libs\":\"['scipy', 'heapq']\"}","display_format":"code","language":"","answer_status":"published","assets":[],"source_url":"https://bigcode-bench.github.io/","history":"initial import","indexing_mode":"index","subproblems":[],"grids":[]}