def test_train_test_split():
num_samples = 40
seed = random.randint(1, 100)
test_size = 0.714
num_train = int(num_samples * (1 - test_size))
num_test = num_samples - num_train
a_shape = (num_samples, 3, 5)
a = np.random.random(a_shape)
b_shape = (num_samples, 6, 9)
b = np.random.random(b_shape)
train_a, test_a, train_b, test_b = train_test_split(a, b, test_size=test_size, seed=seed)
assert train_a.shape == (num_train, 3, 5)
assert train_b.shape == (num_train, 6, 9)
assert len(train_a) == len(train_b)
assert test_a.shape == (num_test, 3, 5)
assert test_b.shape == (num_test, 6, 9)
assert len(test_a) == len(test_b)
same_seed_train_a, same_seed_test_a, same_seed_train_b, same_seed_test_b = \
train_test_split(a, b, test_size=test_size, seed=seed)
assert np.all(train_a == same_seed_train_a)
assert np.all(train_b == same_seed_train_b)
assert np.all(test_a == same_seed_test_a)
assert np.all(test_b == same_seed_test_b)
diff_seed_train_a, diff_seed_test_a, diff_seed_train_b, diff_seed_test_b = \
train_test_split(a, b, test_size=test_size, seed=0)
assert not np.all(train_a == diff_seed_train_a)
assert not np.all(train_b == diff_seed_train_b)
assert not np.all(test_a == diff_seed_test_a)
assert not np.all(test_b == diff_seed_test_b)
def files(self):
all_image_files = self._all_files()
if self.validation_size > 0:
train_image_files, test_image_files = train_test_split(
all_image_files, test_size=self.validation_size, seed=1)
if self.subset == "train":
files = train_image_files
else:
files = test_image_files
return files
return all_image_files
def files_and_annotations(self):
files, annotations = self._files_and_annotations()
if self.validation_size > 0:
train_files, test_files, train_annotations, test_annotations = train_test_split(
files, annotations, test_size=self.validation_size, seed=1)
if self.subset == "train":
files = train_files
annotations = train_annotations
if self.subset == "validation":
files = test_files
annotations = test_annotations
files = files
annotations = annotations
return files, annotations
def files_and_annotations(self):
"""Return image and annotation file list.
If there is no test dataset, then split dataset to train and test lists with specific ratio.
"""
if self.subset == "train" or self.validation_size > 0:
text = "train.txt"
else:
text = "val.txt"
filename = os.path.join(self.data_dir, text)
df = pd.read_csv(
filename,
delim_whitespace=True,
header=None,
names=['image_files', 'label_files'],
)
image_files = df.image_files.tolist()
label_files = df.label_files.tolist()
image_files = [
os.path.join(self.data_dir, filename) for filename in image_files
]
label_files = [
os.path.join(self.data_dir, filename) for filename in label_files
]
if self.validation_size > 0:
train_image_files, test_image_files, train_label_files, test_label_files = \
train_test_split(image_files, label_files, test_size=self.validation_size, seed=1)
if self.subset == "train":
image_files = train_image_files
label_files = train_label_files
else:
image_files = test_image_files
label_files = test_label_files
image_files, label_files = shuffle(image_files, label_files)
print("files and annotations are ready")
return image_files, label_files
def files_and_annotations(self):
"""Return image and annotation file list.
If there is no test dataset, then split dataset to train and test lists with specific ratio.
"""
if self.subset == "train" or self.validation_size > 0:
text = "train.txt"
else:
text = "val.txt"
filename = os.path.join(self.data_dir, text)
image_files, label_files = list(), list()
with open(filename) as f:
for line in f:
items = line.split()
image_files.append(items[0])
label_files.append(items[1])
image_files = [
os.path.join(self.data_dir, filename) for filename in image_files
]
label_files = [
os.path.join(self.data_dir, filename) for filename in label_files
]
if self.validation_size > 0:
train_image_files, test_image_files, train_label_files, test_label_files = \
train_test_split(image_files, label_files, test_size=self.validation_size, seed=1)
if self.subset == "train":
image_files = train_image_files
label_files = train_label_files
else:
image_files = test_image_files
label_files = test_label_files
image_files, label_files = shuffle(image_files, label_files)
print("files and annotations are ready")
return image_files, label_files
def files_and_annotations(self):
"""Return all files and labels list."""
all_files, all_annotations = self._files_and_annotations(
self.path['json'], self.path['dir'])
if self.validation_size > 0:
train_files, test_files, train_annotations, test_annotations =\
train_test_split(all_files,
all_annotations,
test_size=self.validation_size,
seed=1)
if self.subset == "train":
files = train_files
annotations = train_annotations
else:
files = test_files
annotations = test_annotations
else:
files, annotations = all_files, all_annotations
return files, annotations