alibox.split_AL(test_ratio=0.3, initial_label_rate=0.05, split_count=5)
# Use the default Logistic Regression classifier
model = alibox.get_default_model()
# The cost budget is 50 times querying
stopping_criterion = alibox.get_stopping_criterion('num_of_queries', 30)
# experiment
meta_regressor = joblib.load('meta_lr.joblib')
# meta_query = QueryMetaData(X, y, meta_regressor)
meta_result = []
for round in range(5):
meta_query = QueryMetaData(X, y, meta_regressor)
# Get the data split of one fold experiment
train_idx, test_idx, label_ind, unlab_ind = alibox.get_split(round)
# Get intermediate results saver for one fold experiment
saver = alibox.get_stateio(round)
# calc the initial point
model.fit(X=X[label_ind.index, :], y=y[label_ind.index])
pred = model.predict(X[test_idx, :])
accuracy = sum(pred == y[test_idx]) / len(test_idx)
saver.set_initial_point(accuracy)
while not stopping_criterion.is_stop():
# Select a subset of Uind according to the query strategy
# Passing model=None to use the default model for evaluating the committees' disagreement
select_ind = meta_query.select(label_ind, unlab_ind, model=None)
label_ind.update(select_ind)
# Use pre-defined strategy
unc = QueryInstanceUncertainty(X, y)
qbc = QueryInstanceQBC(X, y)
eer = QureyExpectedErrorReduction(X, y)
# random = QueryRandom(X, y)
unc_result.append(copy.deepcopy(main_loop(alibox, unc, round)))
qbc_result.append(copy.deepcopy(main_loop(alibox, qbc, round)))
# eer_result.append(copy.deepcopy(main_loop(alibox, eer, round)))
# random_result.append(copy.deepcopy(main_loop(alibox, random, round)))
rfr_regression_result = []
for round in range(splitcount):
meta_query = QueryMetaData(X, y, rfr)
# Get the data split of one fold experiment
train_idx, test_idx, label_ind, unlab_ind = alibox.get_split(round)
# Get intermediate results saver for one fold experiment
saver = alibox.get_stateio(round)
# calc the initial point
model.fit(X=X[label_ind.index, :], y=y[label_ind.index])
pred = model.predict(X[test_idx, :])
accuracy = sum(pred == y[test_idx]) / len(test_idx)
saver.set_initial_point(accuracy)
while not stopping_criterion.is_stop():
# Select a subset of Uind according to the query strategy
# Passing model=None to use the default model for evaluating the committees' disagreement
select_ind = meta_query.select(label_ind, unlab_ind, model=model)
label_ind.update(select_ind)
s_c_300_1model_result = []
q_t_50_1model_result = []
q_t_70_1model_result = []
q_t_90_1model_result = []
q_t_100_1model_result = []
q_t_110_1model_result = []
q_t_200_1model_result = []
q_t_300_1model_result = []
ridge_qt50_7_result = []
for name, regressor in meta_regressor.items():
for round in range(splitcount):
meta_query = QueryMetaData(X, y, regressor, copy.deepcopy(label_index_round[round]), copy.deepcopy(unlabel_index_round[round]), copy.deepcopy(model_output_round[round]))
# Get the data split of one fold experiment
train_idx, test_idx, label_ind, unlab_ind = alibox.get_split(round)
# Get intermediate results saver for one fold experiment
saver = alibox.get_stateio(round)
# calc the initial point
label_ind = copy.deepcopy(label_index_round[round][4])
unlab_ind = copy.deepcopy(unlabel_index_round[round][4])
model.fit(X=X[label_ind.index, :], y=y[label_ind.index])
pred = model.predict(X[test_idx, :])
accuracy = sum(pred == y[test_idx]) / len(test_idx)
saver.set_initial_point(accuracy)
while not stopping_criterion.is_stop():
# Select a subset of Uind according to the query strategy
# Passing model=None to use the default model for evaluating the committees' disagreement
# st = alibox.State(select_index=select_ind, performance=accuracy)
# saver.add_state(st)
# saver.save()
# # Passing the current progress to stopping criterion object
# stopping_criterion.update_information(saver)
# # Reset the progress in stopping criterion object
# stopping_criterion.reset()
# lr_cdata_random_result.append(copy.deepcopy(saver))
# the regression way
rfr_regression_result = []
for round in range(splitcount):
meta_query = QueryMetaData(X, y, rfr_meta,
copy.deepcopy(label_index_round[round]),
copy.deepcopy(unlabel_index_round[round]),
copy.deepcopy(model_output_round[round]))
# Get the data split of one fold experiment
train_idx, test_idx, label_ind, unlab_ind = alibox.get_split(round)
# Get intermediate results saver for one fold experiment
saver = alibox.get_stateio(round)
# calc the initial point
label_ind = copy.deepcopy(label_index_round[round][4])
unlab_ind = copy.deepcopy(unlabel_index_round[round][4])
model.fit(X=X[label_ind.index, :], y=y[label_ind.index])
pred = model.predict(X[test_idx, :])
accuracy = sum(pred == y[test_idx]) / len(test_idx)
saver.set_initial_point(accuracy)
while not stopping_criterion.is_stop():
# Select a subset of Uind according to the query strategy