{"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":"414943b0-e237-5079-bd5d-2177351db1ff","task_key":"default--v0~2e1~2e0~5fhf--414943b0-e237-5079-bd5d-2177351db1ff","task_revision_id":"1","upstream_id":"","short_description":"BigCodeBench v0.1.0_hf 414943b0-e237-5079-bd5d-2177351db1ff","config":"default","split":"v0.1.0_hf","body":"{\"code_prompt\":\"import json\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nimport pandas as pd\\nimport numpy as np\\nfrom collections import defaultdict\\ndef task_func(input_file: str) -> plt.Axes:\\n\",\"complete_prompt\":\"import json\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nimport pandas as pd\\nimport numpy as np\\nfrom collections import defaultdict\\n\\n\\ndef task_func(input_file: str) -> plt.Axes:\\n    \\\"\\\"\\\"\\n    Read a list of dictionaries from a JSON file, calculate the results (mean and median for each key)\\n    via numpy, convert the input data into a pandas DataFrame with the keys as \\\"X\\\" and values as \\\"Y\\\"\\n    for visualization with a seaborn box plot, then return the results and box plot.\\n\\n    Parameters:\\n    - input_file (str): The input JSON file name with absolute path.\\n\\n    Returns:\\n    - results (dict): Dictionary where each key is a unique key from the original input, and each\\n                      value is a corresponding dict, with keys 'mean' and 'median' and the statistics\\n                      as values.\\n    - ax (plt.Axes): The box plot of aggregated 'Values for Each Key' in the input data.\\n\\n    Requirements:\\n    - json\\n    - seaborn\\n    - matplotlib.pyplot\\n    - pandas\\n    - numpy\\n    - collections.defaultdict\\n\\n    Example:\\n    >>> results, ax = task_func(\\\"/path/to/data.json\\\")\\n    >>> ax\\n    <class 'matplotlib.axes._axes.Axes'>\\n    >>> results\\n    {'a': {'mean': 3.0, 'median': 3.0}, 'b': {'mean': 2.0, 'median': 3.0}}\\n    \\\"\\\"\\\"\\n\",\"entry_point\":\"task_func\",\"instruct_prompt\":\"Read a list of dictionaries from a JSON file, calculate the results (mean and median for each key) via numpy, convert the input data into a pandas DataFrame with the keys as \\\"X\\\" and values as \\\"Y\\\" for visualization with a seaborn box plot, then return the results and box plot.\\nThe function should output with:\\n    results (dict): Dictionary where each key is a unique key from the original input, and each\\n    value is a corresponding dict, with keys 'mean' and 'median' and the statistics\\n    as values.\\n    ax (plt.Axes): The box plot of aggregated 'Values for Each Key' in the input data.\\nYou should write self-contained code starting with:\\n```\\nimport json\\nimport seaborn as sns\\nimport matplotlib.pyplot as plt\\nimport pandas as pd\\nimport numpy as np\\nfrom collections import defaultdict\\ndef task_func(input_file: str) -> plt.Axes:\\n```\",\"libs\":\"['pandas', 'collections', 'matplotlib', 'numpy', 'seaborn', 'json']\"}","display_format":"code","language":"","answer_status":"published","assets":[],"source_url":"https://bigcode-bench.github.io/","history":"initial import","indexing_mode":"index","subproblems":[],"grids":[]}