'''
from evaluation.experiment import Experiment
import data.load_data as load_data
import numpy as np
output_dir = '../../data/bayesian_sequence_combination/output/ner/'
regen_data = False
gt, annos, doc_start, text, gt_nocrowd, doc_start_nocrowd, text_nocrowd, gt_task1_val, gt_val, doc_start_val, text_val, _ = \
load_data.load_ner_data(regen_data)
# ------------------------------------------------------------------------------------------------
exp = Experiment(None, 9, annos.shape[1], None, max_iter=20)
exp.save_results = True
exp.opt_hyper = False #True
best_bac_wm = 'bac_seq' #'unknown' # choose model with best score for the different BAC worker models
best_bac_wm_score = -np.inf
best_nu0factor = 0.1
best_diags = 1
best_factor = 1
best_acc_bias = 0
exp.alpha0_diags = best_diags
exp.alpha0_factor = best_factor
exp.nu0_factor = best_nu0factor
exp.alpha0_acc_bias = best_acc_bias
exp.methods = [
annos,
gt,
doc_start,
features,
annos,
gt_val,
doc_start,
features,
alpha0_factor=alpha0_factor,
alpha0_diags=alpha0_diags,
beta0_factor=beta0_factor,
max_iter=20)
exp.methods = [
'ibcc',
]
exp.opt_hyper = False
exp.run_methods(new_data=regen_data)
# ----------------------------------------------------------------------------
beta0_factor = 0.1
alpha0_diags = 0.1
alpha0_factor = 0.1
output_dir = os.path.join(
evaluation.experiment.output_root_dir,
'ner3_%f_%f_%f' % (beta0_factor, alpha0_diags, alpha0_factor))
exp = Experiment(output_dir,
9,
annos,
gt,
doc_start,
