Forman Quantification Datasets (without splits)

The datasets were created using the same protocol and data provided and established in [1].

For future experiment, we maintain all data together in order to allow each resercher to maintain their train/test split controlled.

To use and reproduce the experiments, please cite the original paper that provided and prepared the experimental protocol from these splits [1].

[1] Forman, G. (2005). Counting Positives Accurately Despite Inaccurate Classification. In: Gama, J., Camacho, R., Brazdil, P.B., Jorge, A.M., Torgo, L. (eds) Machine Learning: ECML 2005. ECML 2005. Lecture Notes in Computer Science, vol. 3720. Springer, Berlin, Heidelberg.

In order to read the data, you can use the following code

import numpy as np
import os
import scipy.sparse as sp

X_train = sp.load_npz(os.path.join(...))
y_train = np.load(os.path.join(...), allow_pickle=True)