Calculate uncertainty measures from Monte Carlo sampled model outputs.
Project description
MC Uncertainty
Uncertainty estimation functions for use with Monte Carlo sampled model outputs.
Installation
pip install mc-uncertainty
Usage
import mc_uncertainty as mcu
# All functions accept data with shape (mc_samples, n, classes)
data = np.array(...
# Variance
print(mcu.variance(data).shape) # [n, classes]
# Entropy
print(mcu.entropy(data).shape) # [mc_samples, n]
# Predicted entropy
print(mcu.predicted_entropy(data).shape) # [n,]
# Mutual information
print(mcu.mutual_information(data).shape) # [n,]
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
mc-uncertainty-0.1.3.tar.gz
(1.5 kB
view hashes)