Students who need it can contact them via private message. It is recommended to pay attention to the subscription account platform in the lower right corner of the above picture to download it.
Standard time series data will appear in time series papers in top journals such as ICML, IEEE, and CVPR, and data sets in fields such as electric power, transportation, and new energy are included. Today we will briefly summarize and share the time series data sets that are frequently used in the top papers. Here are 4 data sets that can be used for research related to multi-dimensional time series signals. The format of different data sets is the same. Assuming a time series signal contains T timestamps and n features for each timestamp, the data file will contain T lines, each line having n real numbers separated by commas. Come on, show~
01
—
Electricity data set
This is a standard dataset from UCI, the original dataset is at https://archive.ics.uci.edu/ml/datasets/ElectricityLoadDiagrams20112014. It contains electricity consumption recorded every 15 minutes from 2011 to 2014. Because some dimensions are equal to 0, the records in 2011 are excluded from the data set. The final data we obtained was the electricity usage of 370 customers during the time period from 2012 to 2014, and we converted the data to reflect hourly consumption.
02
—
Traffic traffic data set
The original data is located at http://pems.dot.ca.gov, and the data in this dataset is a collection of hourly data from the California Department of Transportation for 48 consecutive months (2015-2016). This data describes road occupancy (between 0 and 1) measured by different sensors on San Francisco Bay Area highways.
03
—
Solar data set
The original data is located at http://www.nrel.gov/grid/solar-power-data.html. This data set is relatively large. It contains the solar power generation record data of each major state in the United States in 2006. The photovoltaic power generation record data is collected every 10 minutes. The power station samples the status once. Additionally, NREL Transforming Energy's Integrated Research Solar Data represents U.S. synthetic solar photovoltaic (PV) power plant data points in 2006. The data are used by energy professionals such as transmission planners, utility planners, project developers and university researchers who conduct solar energy integration studies and need to estimate the power generation of hypothetical solar plants.
04
—
Exchange rate exchange_rate data set
The exchange rate standard data set contains daily exchange rate summaries from 1990 to 2016 for eight countries including Australia, the United Kingdom, Canada, Switzerland, China, Japan, New Zealand, and Singapore, with a data granularity of 1D.
05
—
Conclusion
The above is all the contents of the standard data set for time series papers. To download the data set, please pay attention to the platform in the lower right corner of the article picture to obtain it.