This Python package implements our unified framework of algorithms for quantification and unfolding. It is designed for enabling the composition of novel methods from existing and easily customized loss functions and data representations. Moreover, this package leverages a powerful optimization back-end to yield state-of-the-art performances for all compositions.
pip install --upgrade pip setuptools wheel
pip install 'qunfold @ git+https://github.com/mirkobunse/qunfold'
Moreover, you will need a JAX backend. Typically, the CPU backend will be ideal:
pip install "jax[cpu]"
For detailed information, visit the documentation.
Basically, you use this package as follows:
from qunfold import ACC # Adjusted Classify and Count
from sklearn.ensemble import RandomForestClassifier
acc = ACC( # use OOB predictions for training the quantifier
RandomForestClassifier(oob_score=True)
)
acc.fit(X_trn, y_trn) # fit to training data
p_hat = acc.predict(X_tst) # estimate a prevalence vector