The dataset contains IEEE 30-bus, 118-bus, 300-bus dataset we generated for learning for unit commitment. The dataset consists of data from both normal and extended time scales, with a total time span of one year. A data point is defined as the load demand for each period and the on/off status of the units at that moment. This dataset can be used to train a neural network to learn the mapping from load demand information to the on/off status of the units.