Datasets for training/testing Machine Learning algorithms. The original source/s of the data is/are mentioned in the dataset folder. What I mostly do here is some preprocessing to make the data suitable for my algorithms or purposes(dividing it into training and test data for instance).
Datasets are organized by algorithm.
Inside each dataset folder you will find a README file with information about the dataset, the original dataset and the preprocessed one. you might find as well a Jupyter notebook showing how the data was preprocessed, a results file with the results obtained for certain algorithm and parameters.