def load_data(self):
"""
Load data from Purchase100 dataset
# Returns
all_data : train data, train labels, test data and test labels
"""
path_features = get_file(
"purchase100",
origin=
"https://github.com/xehartnort/Purchase100-dataset/releases/download/v1.1/purchase100.npz",
extract=True,
file_hash="0d7538b9806e7ee622e1a252585e7768", # md5 hash
cache_dir='~/.sherpa-ai')
all_data = np.load(path_features)
data = all_data['features']
labels = to_categorical(all_data['labels'])
test_size = int(len(data) * 0.1)
self._train_data, self._train_labels,\
self._test_data, self._test_labels = db.split_train_test(
data, labels, test_size)
self.shuffle()
return self.data
def load_data(self):
"""
Load data from iris package
# Returns
all_data : train data, train labels, test data and test labels
"""
all_data = sklearn.datasets.load_iris()
data = all_data["data"]
labels = all_data["target"]
test_size = int(len(data) * 0.1)
self._train_data, self._train_labels,\
self._test_data, self._test_labels = db.split_train_test(data, labels, test_size)
self.shuffle()
return self.data
def load_data(self):
"""
Load data from lfw package
# Returns
all_data : train data, train labels, test data and test labels
"""
all_data = fetch_lfw_people()
data = all_data["data"]
labels = all_data["target"]
test_size = int(len(data) * 0.1)
self._train_data, self._train_labels,\
self._test_data, self._test_labels = db.split_train_test(data, labels, test_size)
self.shuffle()
return self.data