def _download(self):
if not os.path.exists(os.path.join(self.data_path, "CUB_200_2011")):
tgz_path = os.path.join(self.data_path, "CUB_200_2011.tgz")
if not os.path.exists(tgz_path):
print("Downloading tgz archive...", end=' ')
download_file_from_google_drive(
"1hbzc_P1FuxMkcabkgn9ZKinBwW683j45", tgz_path)
print('Done!')
print('Extracting archive...', end=' ')
try:
untar(tgz_path)
except tarfile.ReadError:
print()
try:
os.remove(tgz_path)
except:
pass
raise IOError(
"The file wasn't downloaded properly, probably because it "
"has hit the limit of download from Google Drive. "
"Try downloading the .tgz archive manually at "
"http://www.vision.caltech.edu/visipedia-data/CUB-200/images.tgz"
)
print('Done!')
def _download(self):
if not os.path.exists(os.path.join(self.data_path, "images")):
archive_images_path = os.path.join(self.data_path, "images.tar.gz")
if not os.path.exists(archive_images_path):
print("Downloading images archive...", end=' ')
image_url = os.path.join(self.base_url, "images.tar.gz")
download(image_url, self.data_path)
print('Done!')
print('Extracting images archive...', end=' ')
untar(archive_images_path)
print('Done!')
if not os.path.exists(os.path.join(self.data_path, "annotations")):
archive_annotations_path = os.path.join(self.data_path, "annotations.tar.gz")
if not os.path.exists(archive_annotations_path):
print("Downloading annotations archive...", end=' ')
annotations_url = os.path.join(self.base_url, "annotations.tar.gz")
download(annotations_url, self.data_path)
print('Done!')
print('Extracting annotations archive...', end=' ')
untar(archive_annotations_path)
print('Done!')
def _download(self):
# Downloading images
if not os.path.exists(os.path.join(self.data_path, "VOCdevkit")):
path = os.path.join(self.data_path, "VOCtrainval_11-May-2012.tar")
if not os.path.exists(path):
print("Downloading Pascal VOC segmentation maps...")
download.download(self.data_url, self.data_path)
print("Uncompressing images...")
download.untar(path)
# Downloading segmentation maps
if not os.path.exists(os.path.join(self.data_path, "SegmentationClassAug")):
path = os.path.join(self.data_path, "SegmentationClassAug.zip")
if not os.path.exists(path):
print("Downloading Pascal VOC segmentation maps...")
download.download(self.segmentation_url, self.data_path)
print("Uncompressing segmentation maps...")
download.unzip(path)
# Downloading train/val/test indexes
if not os.path.exists(os.path.join(self.data_path, "list")):
path = os.path.join(self.data_path, "list.zip")
if not os.path.exists(path):
print("Downloading Pascal VOC train/val/test indexes...")
download.download(self.split_url, self.data_path)
print("Uncompressing train/val/test indexes...")
download.unzip(path)
def _download(self):
archive_path = os.path.join(self.data_path, "dtd-r1.0.1.tar.gz")
if not os.path.exists(archive_path):
print("Downloading DTD dataset...")
download.download(self.url, self.data_path)
if not os.path.exists(os.path.join(self.data_path, "dtd")):
print("Uncompressing images...")
download.untar(archive_path)
def _download(self):
archive_path = os.path.join(self.data_path, "fer2013.tar.gz")
if not os.path.exists(archive_path):
raise Exception(
"You need to download this dataset yourself at "
"https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge/data?select=fer2013.tar.gz"
)
if not os.path.exists(os.path.join(self.data_path, "fer2013")):
print("Extracting archive...", end=" ")
untar(archive_path)
print("Done!")
def _download(self):
if not os.path.exists(os.path.join(self.data_path, "VLCS")):
tar_path = os.path.join(self.data_path, "VLCS.tar.gz")
if not os.path.exists(tar_path):
print("Downloading zip images archive...", end=' ')
download_file_from_google_drive(self.images_gdrive_id, tar_path)
print('Done!')
print('Extracting archive...', end=' ')
untar(tar_path)
print('Done!')
def _download(self):
if not os.path.exists(os.path.join(self.data_path, "SUN397")):
archive_path = os.path.join(self.data_path, "SUN397.tar.gz")
if not os.path.exists(archive_path):
print("Downloading images archive...", end=' ')
download(self.images_url, self.data_path)
print('Done!')
print('Extracting archive...', end=' ')
untar(archive_path)
print('Done!')
def _download(self):
if not os.path.exists(
os.path.join(self.data_path, "fgvc-aircraft-2013b")):
archive_path = os.path.join(self.data_path,
"fgvc-aircraft-2013b.tar.gz")
if not os.path.exists(archive_path):
print(f"Downloading archive ...", end=" ")
download(self.url, self.data_path)
print('Done!')
print(f"Extracting archive...", end=" ")
untar(archive_path)
print("Done!")
def _download(self):
data_folder = os.path.join(self.data_path, self.folder)
annotation_folder = os.path.join(self.data_path, "Annotations")
if not os.path.exists(data_folder):
if not os.path.exists(data_folder + ".tar.gz"):
print("Downloading data archive...", end=" ")
download_file_from_google_drive(self.data_id,
data_folder + ".tar.gz")
print("Done!")
print("Extracting data archive...", end=" ")
untar(data_folder + ".tar.gz")
print("Done!")
def _download(self):
if not os.path.exists(
os.path.join(self.data_path, "VOCdevkit", "VOC2007")):
archive_path = os.path.join(self.data_path,
"VOCtrainval_06-Nov-2007.tar")
if not os.path.exists(archive_path):
print(f"Downloading archive ...", end=" ")
download.download(self.url, self.data_path)
print('Done!')
print(f"Extracting archive...", end=" ")
download.untar(archive_path)
print("Done!")
def _download(self):
folders = ["devkit", "cars_train", "cars_test"]
archives = ["car_devkit.tgz", "cars_train.tgz", "cars_test.tgz"]
urls = [self.devkit_url, self.train_url, self.test_url]
for f, a, u in zip(folders, archives, urls):
if not os.path.exists(os.path.join(self.data_path, f)):
archive_path = os.path.join(self.data_path, a)
if not os.path.exists(archive_path):
print(f"Downloading archive {a} ...", end=" ")
download(u, self.data_path)
print('Done!')
print(f"Extracting archive... {a}->{f}", end=" ")
untar(archive_path)
print("Done!")
if not os.path.exists(
os.path.join(self.data_path,
"cars_test_annos_withlabels.mat")):
download(self.test_labels_url, self.data_path)
def _download(self):
if not os.path.exists(os.path.join(self.data_path, "jpg")):
archive_images_path = os.path.join(self.data_path, "102flowers.tgz")
if not os.path.exists(archive_images_path):
print("Downloading images archive...", end=' ')
image_url = os.path.join(self.base_url, "102flowers.tgz")
download(image_url, self.data_path)
print('Done!')
print('Extracting archive...', end=' ')
untar(archive_images_path)
print('Done!')
# Downloading label file
if not os.path.exists(os.path.join(self.data_path, "imagelabels.mat")):
label_url = os.path.join(self.base_url, "imagelabels.mat")
download(label_url, self.data_path)
# Downloading split file
if not os.path.exists(os.path.join(self.data_path, "setid.mat")):
split_url = os.path.join(self.base_url, "setid.mat")
download(split_url, self.data_path)
def _download(self):
global _DOWNLOAD_FOLDER
_DOWNLOAD_FOLDER = os.path.join(self.data_path, "images")
os.makedirs(_DOWNLOAD_FOLDER, exist_ok=True)
if not os.path.exists(os.path.join(self.data_path, "birdsnap")):
archive_path = os.path.join(self.data_path, "birdsnap.tgz")
if not os.path.exists(archive_path):
print("Downloading archive of metadata...", end=' ')
download(self.meta_url, self.data_path)
print('Done!')
print('Extracting archive...', end=' ')
untar(archive_path)
print('Done!')
with open(os.path.join(self.data_path, "birdsnap", "images.txt")) as f:
data = f.readlines()[1:]
good_images = 0
print(f"Downloading or checking {len(data)} images...")
pb = ProgressBar()
with ThreadPool(10) as pool:
for processed_data in pool.imap_unordered(_download_images, data):
pb.update(None, 1, len(data))
if processed_data is None:
continue
good_images += 1
pb.end(len(data))
if good_images != len(data):
warnings.warn(
f"{len(data)-good_images} couldn't be downloaded among {len(data)}."
)
def _download(self):
if not os.path.exists(
os.path.join(self.data_path, "eccv_18_all_images_sm")):
tar_path = os.path.join(self.data_path,
"eccv_18_all_images_sm.tar.gz")
if not os.path.exists(tar_path):
print("Downloading images archive...", end=" ")
download(self.images_url, self.data_path)
print("Done!")
print('Extracting archive...', end=' ')
untar(tar_path)
print('Done!')
if not os.path.exists(
os.path.join(self.data_path, "caltech_images_20210113.json")):
zip_path = os.path.join(self.data_path,
"caltech_camera_traps.json.zip")
if not os.path.exists(zip_path):
print("Downloading json archive...", end=" ")
download(self.json_url, self.data_path)
print("Done!")
print('Extracting archive...', end=' ')
unzip(zip_path)
print('Done!')
