{"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":"5f2112d6-0f7e-5cdd-887b-0ab140f49178","task_key":"default--v0~2e1~2e0~5fhf--5f2112d6-0f7e-5cdd-887b-0ab140f49178","task_revision_id":"1","upstream_id":"","short_description":"BigCodeBench v0.1.0_hf 5f2112d6-0f7e-5cdd-887b-0ab140f49178","config":"default","split":"v0.1.0_hf","body":"{\"code_prompt\":\"import math\\nimport numpy as np\\nfrom datetime import datetime\\nimport pandas as pd\\ndef task_func(\\n    start_time,\\n    end_time,\\n    step,\\n    columns=[\\\"Timestamp\\\", \\\"Sensor1\\\", \\\"Sensor2\\\", \\\"Sensor3\\\", \\\"SensorStatus\\\"],\\n    sensor_statuses=[\\\"OK\\\", \\\"MAINTENANCE_REQUIRED\\\", \\\"ERROR\\\"],\\n    random_seed=42,\\n):\\n\",\"complete_prompt\":\"import math\\nimport numpy as np\\nfrom datetime import datetime\\nimport pandas as pd\\n\\n\\ndef task_func(\\n    start_time,\\n    end_time,\\n    step,\\n    columns=[\\\"Timestamp\\\", \\\"Sensor1\\\", \\\"Sensor2\\\", \\\"Sensor3\\\", \\\"SensorStatus\\\"],\\n    sensor_statuses=[\\\"OK\\\", \\\"MAINTENANCE_REQUIRED\\\", \\\"ERROR\\\"],\\n    random_seed=42,\\n):\\n    \\\"\\\"\\\"\\n    Generate a DataFrame with detailed artificial sensor readings for specified timestamps\\n    and sensor statuses from a predefined list.\\n\\n    The function generates sensor readings for Sensor1, Sensor2, and Sensor3 (or their\\n    corresponding named columns in the supplied column list) using sine, cosine, and tan\\n    functions, respectively, of the timestamp (converted to seconds), with a small random\\n    noise added to simulate real sensor data variability.\\n    SensorStatus is randomly chosen from the provided statuses for each timestamp.\\n\\n    Parameters:\\n    - start_time (int): Start time in milliseconds since epoch.\\n    - end_time (int): End time in milliseconds since epoch. Must not be before start_time.\\n    - step (int): The interval in milliseconds between each generated data point. Must be positive.\\n                  This step defines the frequency at which data points are generated. If the step\\n                  does not neatly divide the interval between start_time and end_time into\\n                  equal-sized portions, the last timestamp may be excluded.\\n    - columns (list of str, optional): Names of the DataFrame columns to be included in the output.\\n                                       Defaults to: ['Timestamp', 'Sensor1', 'Sensor2', 'Sensor3', 'SensorStatus'].\\n                                       Regardless of naming, the function will populate the first column with\\n                                       timestamp, the middle columns with sensor data, and the final with status.\\n    - sensor_statuses (list of str, optional): Possible statuses for the sensors to randomly assign in the dataset.\\n                                               Defaults to: ['OK', 'MAINTENANCE_REQUIRED', 'ERROR'].\\n    - random_seed (int, optional): Seed for the random number generator to ensure reproducible results.\\n                                   Defaults to 42.\\n\\n    Returns:\\n    - pd.DataFrame: Generated sensor readings for the given timestamps.\\n\\n    Requirements:\\n    - math\\n    - datetime\\n    - numpy\\n    - pandas\\n\\n    Example:\\n    >>> df = task_func(0, 5000, 1000)\\n    >>> type(df)\\n    <class 'pandas.core.frame.DataFrame'>\\n    >>> df.head(1)\\n                        Timestamp   Sensor1   Sensor2   Sensor3 SensorStatus\\n    0  1970-01-01 00:00:00.000000  0.049671  0.986174  0.064769        ERROR\\n    \\\"\\\"\\\"\\n\",\"entry_point\":\"task_func\",\"instruct_prompt\":\"Generate a DataFrame with detailed artificial sensor readings for specified timestamps and sensor statuses from a predefined list. The function generates sensor readings for Sensor1, Sensor2, and Sensor3 (or their corresponding named columns in the supplied column list) using sine, cosine, and tan functions, respectively, of the timestamp (converted to seconds), with a small random noise added to simulate real sensor data variability. SensorStatus is randomly chosen from the provided statuses for each timestamp.\\nThe function should output with:\\n    pd.DataFrame: Generated sensor readings for the given timestamps.\\nYou should write self-contained code starting with:\\n```\\nimport math\\nimport numpy as np\\nfrom datetime import datetime\\nimport pandas as pd\\ndef task_func(\\n    start_time,\\n    end_time,\\n    step,\\n    columns=[\\\"Timestamp\\\", \\\"Sensor1\\\", \\\"Sensor2\\\", \\\"Sensor3\\\", \\\"SensorStatus\\\"],\\n    sensor_statuses=[\\\"OK\\\", \\\"MAINTENANCE_REQUIRED\\\", \\\"ERROR\\\"],\\n    random_seed=42,\\n):\\n```\",\"libs\":\"['math', 'pandas', 'datetime', 'numpy']\"}","display_format":"code","language":"","answer_status":"published","assets":[],"source_url":"https://bigcode-bench.github.io/","history":"initial import","indexing_mode":"index","subproblems":[],"grids":[]}