You can download the data set here: link
The data will be divided to the training sets and the test sets. The training data is composed of two files. train.csv contains room reservation requests. The meanings of each field can be found in the original Kaggle site. train label.csv contains the revenue “truth” for each day of the training period.
The test data is essentially the same as the training one. test.csv is the counterpart of train.csv, with some “future” fields hidden. test_nolabel.csv is the counterpart of train_label.csv, except that the “truth” is hidden. You are not allowed to peep/label the true answers of the test sets.The goal is estimate the quantized label of daily revenue, as precisely as possible, given the booking records in the future.
There are 32 features and 10 quantized labels for each testcase. Please try and conqure the task.