lambdaU = 1. / 10.
lambdaV = 1. / 10.
priors = {'alpha': alpha, 'beta': beta, 'lambdaU': lambdaU, 'lambdaV': lambdaV}
# Load in the Sanger dataset
(_, X_min, M, _, _, _, _) = load_Sanger(standardised=standardised)
# Run the cross-validation framework
random.seed(42)
numpy.random.seed(9000)
nested_crossval = LineSearchCrossValidation(classifier=bnmf_vb_optimised,
R=X_min,
M=M,
values_K=K_range,
folds=no_folds,
priors=priors,
init_UV=init_UV,
iterations=iterations,
restarts=restarts,
quality_metric=quality_metric,
file_performance=output_file)
nested_crossval.run()
"""
All model fits for fold 1, metric AIC: [255811.96901730512, 252136.60278205923, 250856.47342754889, 250226.77983334119, 248494.30974350378].
Best K for fold 1: 10.
Performance: {'R^2': 0.8062985415424776, 'MSE': 2.2509307850276188, 'Rp': 0.89865828998196062}.
All model fits for fold 2, metric AIC: [254648.23575785581, 252738.04961239398, 251251.00158825511, 249894.09518978855, 249008.72134400008].
Best K for fold 2: 10.
Performance: {'R^2': 0.8034479975799519, 'MSE': 2.2875509346902207, 'Rp': 0.8971303533111219}.
