# You should have received a copy of the GNU General Public License #
# along with this program. If not, see <http://www.gnu.org/licenses/>. #
# #
# You should have received a copy of the GNU General Public License #
# along with this program. If not, see <http://www.gnu.org/licenses/>. #
# #
# Contact <francois.fleuret@idiap.ch> for comments & bug reports #
#########################################################################
# Contact <francois.fleuret@idiap.ch> for comments & bug reports #
#########################################################################
parser = argparse.ArgumentParser(
description = '''An implementation of a Mutual Information estimator with a deep model
parser = argparse.ArgumentParser(
description = '''An implementation of a Mutual Information estimator with a deep model
- (1) Two MNIST images of same class. The "true" MI is the log of the
- number of used MNIST classes.
+ (1) Two MNIST images of same class. The "true" MI is the log of the
+ number of used MNIST classes.
- (2) One MNIST image and a pair of real numbers whose difference is
- the class of the image. The "true" MI is the log of the number of
- used MNIST classes.
+ (2) One MNIST image and a pair of real numbers whose difference is
+ the class of the image. The "true" MI is the log of the number of
+ used MNIST classes.
- (3) Two 1d sequences, the first with a single peak, the second with
- two peaks, and the height of the peak in the first is the
- difference of timing of the peaks in the second. The "true" MI is
- the log of the number of possible peak heights.''',
+ (3) Two 1d sequences, the first with a single peak, the second with
+ two peaks, and the height of the peak in the first is the
+ difference of timing of the peaks in the second. The "true" MI is
+ the log of the number of possible peak heights.''',