Potential Mismatch Between Sampling Frequency and Time Series Points.

[Liusuthu]

Dear authors,

Thanks for open-sourcing MCQ-1 and MCQ-2 dataset, which is diverse and of high quality.

When checking specific samples, I find maybe there exist potential mismatch between sampling frequency and the number time series points. Take MCQ_1_TS.jsonl and it's first sample for example:

characteristics:  
1) Continuous increment of temperature during power outage, 
2) Slow decrease of temperature once power is restored until it reaches normal levels, 
3) Stable temperature values before the outage and after temperature restoration, 
4) Overall time series has slow ascending and descending trends during the outage and recovery period respectively,
5) Lack of daily temperature variation due to constant artificial control.

description:  A scientist is measuring the temperature in an industrial freezer over two weeks, with a new reading taken every 6 hours. An unexpected power outage occurs in the middle of this period lasting 3 days. This causes the freezer temperature to rise before it is restored and begins dropping again. 

metadata: {'start': '2022-03-01 00:00:00', 'end': '2022-03-15 00:00:00', 'units': 'Degrees Celsius', 'frequency': 'Every 6 hours'}

According to the metadata and description, the sampling frequency is 'every 6 hours' and duration is 2 weeks, which would lead to a series of 56 points. However, the series is if 500 points.

Could you explain it? Is it a factual error?

Best,
Mush Li