def setup(self):
to_uuid = lambda p: os.path.splitext(os.path.basename(p))[0]
data_dir = os.path.join(self.dataset_dir, "data")
if os.path.isdir(data_dir):
self._uuids_to_image_paths = {
to_uuid(p): p
for p in etau.list_files(data_dir, abs_paths=True)
}
else:
self._uuids_to_image_paths = {}
labels_dir = os.path.join(self.dataset_dir, "labels")
if os.path.isdir(labels_dir):
self._uuids_to_labels_paths = {
to_uuid(p): p
for p in etau.list_files(labels_dir, abs_paths=True)
}
else:
self._uuids_to_labels_paths = {}
if self.skip_unlabeled:
uuids = sorted(self._uuids_to_labels_paths.keys())
else:
# Allow uuid to missing from `_uuids_to_image_paths` since we will
# try to use filepath from labels, if present
uuids = sorted(
set(self._uuids_to_image_paths.keys())
| set(self._uuids_to_labels_paths.keys()))
self._uuids = self._preprocess_list(uuids)
self._num_samples = len(self._uuids)
def setup(self):
to_uuid = lambda p: os.path.splitext(os.path.basename(p))[0]
data_dir = os.path.join(self.dataset_dir, "data")
if os.path.isdir(data_dir):
self._uuids_to_video_paths = {
to_uuid(p): p
for p in etau.list_files(data_dir, abs_paths=True)
}
else:
self._uuids_to_video_paths = {}
labels_dir = os.path.join(self.dataset_dir, "labels")
if os.path.isdir(labels_dir):
self._uuids_to_labels_paths = {
to_uuid(p): p
for p in etau.list_files(labels_dir, abs_paths=True)
}
else:
self._uuids_to_labels_paths = {}
if self.skip_unlabeled:
uuids = sorted(self._uuids_to_labels_paths.keys())
else:
uuids = sorted(self._uuids_to_video_paths.keys())
self._info = None
self._uuids = self._preprocess_list(uuids)
self._num_samples = len(self._uuids)
self._cvat_task_labels = CVATTaskLabels()
def setup(self):
data_dir = os.path.join(self.dataset_dir, "data")
labels_dir = os.path.join(self.dataset_dir, "labels")
self._image_paths = etau.list_files(data_dir, abs_paths=True)
self._anno_uuids_to_paths = {
os.path.splitext(f)[0]: os.path.join(labels_dir, f)
for f in etau.list_files(labels_dir, abs_paths=False)
}
def setup(self):
self._sample_parser = FiftyOneImageClassificationSampleParser()
data_dir = os.path.join(self.dataset_dir, "data")
self._image_paths_map = {
os.path.splitext(os.path.basename(p))[0]: p
for p in etau.list_files(data_dir, abs_paths=True)
}
labels_path = os.path.join(self.dataset_dir, "labels.json")
if os.path.isfile(labels_path):
labels = etas.load_json(labels_path)
else:
labels = {}
self._classes = labels.get("classes", None)
self._sample_parser.classes = self._classes
self._labels_map = labels.get("labels", {})
if self.skip_unlabeled:
self._labels_map = {
k: v for k, v in self._labels_map.items() if v is not None
}
uuids = sorted(self._labels_map.keys())
self._uuids = self._preprocess_list(uuids)
self._num_samples = len(self._uuids)
def setup(self):
self._data_dir = os.path.join(self.dataset_dir, "data")
labels_path = os.path.join(self.dataset_dir, "labels.json")
if os.path.isfile(labels_path):
(
info,
classes,
supercategory_map,
images,
annotations,
) = load_coco_detection_annotations(labels_path)
else:
info = {}
classes = None
supercategory_map = None
images = {}
annotations = None
if classes is not None:
info["classes"] = classes
self._info = info
self._classes = classes
self._supercategory_map = supercategory_map
self._images_map = {i["file_name"]: i for i in images.values()}
self._annotations = annotations
if self.skip_unlabeled:
filenames = self._images_map.keys()
else:
filenames = etau.list_files(self._data_dir, abs_paths=False)
self._filenames = self._preprocess_list(filenames)
def setup(self):
filepaths = etau.list_files(
self.dataset_dir, abs_paths=True, recursive=self.recursive
)
filepaths = [p for p in filepaths if etav.is_video_mime_type(p)]
self._filepaths = self._preprocess_list(filepaths)
self._num_samples = len(self._filepaths)
def setup(self):
self._data_dir = os.path.join(self.dataset_dir, "data")
self._labels_path = os.path.join(self.dataset_dir, "labels.xml")
_, cvat_images = load_cvat_image_annotations(self._labels_path)
# Index by filename
self._images_map = {i.name: i for i in cvat_images}
self._filenames = etau.list_files(self._data_dir, abs_paths=False)
def setup(self):
self._data_dir = os.path.join(self.dataset_dir, "data")
labels_path = os.path.join(self.dataset_dir, "labels.json")
classes, images, annotations = load_coco_detection_annotations(
labels_path
)
self._classes = classes
self._images_map = {i["file_name"]: i for i in images.values()}
self._annotations = annotations
self._filenames = etau.list_files(self._data_dir, abs_paths=False)
def setup(self):
self._sample_parser = FiftyOneImageClassificationSampleParser()
data_dir = os.path.join(self.dataset_dir, "data")
self._image_paths_map = {
os.path.splitext(os.path.basename(p))[0]: p
for p in etau.list_files(data_dir, abs_paths=True)
}
labels_path = os.path.join(self.dataset_dir, "labels.json")
labels = etas.load_json(labels_path)
self._sample_parser.classes = labels.get("classes", None)
self._labels = labels.get("labels", {})
self._num_samples = len(self._labels)
def setup(self):
self._data_dir = os.path.join(self.dataset_dir, "data")
self._labels_path = os.path.join(self.dataset_dir, "labels.json")
if os.path.isfile(self._labels_path):
self._anno_dict_map = load_bdd_annotations(self._labels_path)
else:
self._anno_dict_map = {}
filenames = etau.list_files(self._data_dir, abs_paths=False)
if self.skip_unlabeled:
filenames = [f for f in filenames if f in self._anno_dict_map]
self._filenames = self._preprocess_list(filenames)
self._num_samples = len(self._filenames)
def __init__(self, output_dir="", default_ext="", ignore_exts=False):
self.output_dir = output_dir
self.default_ext = default_ext
self.ignore_exts = ignore_exts
self._filename_counts = defaultdict(int)
self._default_filename_patt = (fo.config.default_sequence_idx +
default_ext)
self._idx = 0
if output_dir:
etau.ensure_dir(output_dir)
filenames = etau.list_files(output_dir)
self._idx = len(filenames)
for filename in filenames:
self._filename_counts[filename] += 1
def parse_images_dir(dataset_dir, recursive=True):
"""Parses the contents of the given directory of images.
See :class:`fiftyone.types.dataset_types.ImageDirectory` for format
details. In particular, note that files with non-image MIME types are
omitted.
Args:
dataset_dir: the dataset directory
recursive (True): whether to recursively traverse subdirectories
Returns:
a list of image paths
"""
filepaths = etau.list_files(dataset_dir,
abs_paths=True,
recursive=recursive)
return [p for p in filepaths if etai.is_image_mime_type(p)]
def main(notebook_paths, skip_readme):
if notebook_paths:
notebook_paths = [
os.path.relpath(os.path.abspath(path), ROOT_DIR)
for path in notebook_paths
]
else:
notebook_paths = [
os.path.join("examples", path)
for path in etau.list_files(EXAMPLES_DIR, recursive=True)
if path.endswith(".ipynb") and ".ipynb_checkpoints" not in path
]
for notebook_path in notebook_paths:
_add_notebook_links(notebook_path)
if not skip_readme:
_update_readme()
def setup(self):
self._data_dir = os.path.join(self.dataset_dir, "data")
self._labels_path = os.path.join(self.dataset_dir, "labels.xml")
if os.path.isfile(self._labels_path):
info, _, cvat_images = load_cvat_image_annotations(
self._labels_path)
else:
info = {}
cvat_images = []
self._info = info
# Index by filename
self._images_map = {i.name: i for i in cvat_images}
filenames = etau.list_files(self._data_dir, abs_paths=False)
if self.skip_unlabeled:
filenames = [f for f in filenames if f in self._images_map]
self._filenames = self._preprocess_list(filenames)
self._num_samples = len(self._filenames)
def setup(self):
self._data_dir = os.path.join(self.dataset_dir, "data")
self._labels_path = os.path.join(self.dataset_dir, "labels.json")
self._anno_dict_map = load_bdd_annotations(self._labels_path)
self._filenames = etau.list_files(self._data_dir, abs_paths=False)
def setup(self):
filepaths = etau.list_files(self.dataset_dir,
abs_paths=True,
recursive=self.recursive)
self._filepaths = [p for p in filepaths if etai.is_image_mime_type(p)]
def get_num_samples(dataset_dir):
data_dir = os.path.join(dataset_dir, "data")
if not os.path.isdir(data_dir):
return 0
return len(etau.list_files(data_dir))
