def get_online_obj(algo_name, dataset, data_dir, n_jobs=1):
model_class = model_class_dict[algo_name]
# from test.basic.utils import load_data
sys.path.append('../mindware')
from mindware.datasets.utils import load_data
x, y, _ = load_data(dataset,
data_dir=data_dir,
datanode_returned=False,
preprocess=True,
task_type=0)
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
stratify=y,
random_state=1)
def objective_function(config):
config_dict = config.get_dictionary()
model = model_class(**config_dict, n_jobs=n_jobs, random_state=47)
model.fit(x_train, y_train)
y_pred = model.predict(x_test)
perf = 1 - balanced_accuracy_score(y_test, y_pred)
return perf
return objective_function
def check_datasets(datasets):
for _dataset in datasets:
try:
_ = load_data(_dataset,
data_dir=data_dir,
datanode_returned=False,
preprocess=True,
task_type=0)
except Exception as e:
print(traceback.format_exc())
raise ValueError('Dataset - %s load error!' % _dataset)
def get_meta_learning_configs(X,
y,
task_type,
dataset_name='default',
metric='accuracy',
num_cfgs=5):
if X is None or y is None:
X, y, _ = load_data(dataset_name)
backend = create(temporary_directory=None,
output_directory=None,
delete_tmp_folder_after_terminate=False,
delete_output_folder_after_terminate=False,
shared_mode=True)
dm = XYDataManager(X, y, None, None, task_type, None, dataset_name)
configuration_space = pipeline.get_configuration_space(
dm.info,
include_estimators=None,
exclude_estimators=None,
include_preprocessors=None,
exclude_preprocessors=None)
watcher = StopWatch()
name = os.path.basename(dm.name)
watcher.start_task(name)
def reset_data_manager(max_mem=None):
pass
automlsmbo = AutoMLSMBO(
config_space=configuration_space,
dataset_name=dataset_name,
backend=backend,
total_walltime_limit=1e5,
func_eval_time_limit=1e5,
memory_limit=1e5,
metric=metric,
watcher=watcher,
metadata_directory='components/meta_learning/meta_resource',
num_metalearning_cfgs=num_cfgs)
automlsmbo.reset_data_manager = reset_data_manager
automlsmbo.task = task_type
automlsmbo.datamanager = dm
configs = automlsmbo.get_metalearning_suggestions()
return configs
import os
import sys
import argparse
sys.path.append(os.getcwd())
from mindware.datasets.utils import load_data
parser = argparse.ArgumentParser()
dataset_set = 'diabetes,spectf,credit,ionosphere,lymphography,pc4,' \
'messidor_features,winequality_red,winequality_white,splice,spambase,amazon_employee'
parser.add_argument('--datasets', type=str, default=dataset_set)
args = parser.parse_args()
for dataset in args.datasets.split(','):
raw_data = load_data(dataset, datanode_returned=True)
print(raw_data)
import os
import sys
sys.path.append(os.getcwd())
from autosklearn.smbo import AutoMLSMBO
from autosklearn.constants import *
from autosklearn.data.xy_data_manager import XYDataManager
from autosklearn.util.backend import create
from autosklearn.util import pipeline, StopWatch
from mindware.datasets.utils import load_data
dataset_name = 'diabetes'
X, y, _ = load_data(dataset_name)
def get_meta_learning_configs(X, y, task_type, dataset_name, metric='accuracy', num_cfgs=5):
backend = create(temporary_directory=None,
output_directory=None,
delete_tmp_folder_after_terminate=False,
delete_output_folder_after_terminate=False,
shared_mode=True)
dm = XYDataManager(X, y, None, None, task_type, None, dataset_name)
configuration_space = pipeline.get_configuration_space(dm.info,
include_estimators=None,
exclude_estimators=None,
include_preprocessors=None,
exclude_preprocessors=None)
watcher = StopWatch()
y_valid_pred = model.predict(x_valid)
y_test_pred = model.predict(x_test)
valid_perf = 1 - balanced_accuracy_score(y_valid, y_valid_pred)
test_perf = 1 - balanced_accuracy_score(y_test, y_test_pred)
return valid_perf, test_perf
check_datasets(datasets)
cs = model_class.get_cs()
for dataset in datasets:
# load dataset
x, y, _ = load_data(dataset,
data_dir=data_dir,
datanode_returned=False,
preprocess=True,
task_type=0)
# 6:2:2
x_used, x_test, y_used, y_test = train_test_split(x,
y,
test_size=0.2,
stratify=y,
random_state=1)
x_train, x_valid, y_train, y_valid = train_test_split(x_used,
y_used,
test_size=0.25,
stratify=y_used,
random_state=1)
# make dirs