download_files

class stardust.file.download_files.Downloader(urls, save_path=None)

Bases: object

downloader

Args:

urls:

Required download link

save_path:

Save directory

static is_empty(filename)

async static write_local(result, local_path)

async fetch(session, url, local_path)

async async_download()

stardust.file.download_files.attachment_name_factory(att, save_path, project_id, task_id, frame_num, image_url_lst, pointcloud_url_lst, name=None)

stardust.file.download_files.load_dataset(frame=None, project_id=None, save_path=None, **kwargs)

The function of loading data sets, you can enter the Frame object, you can also enter the project pool, if there is already a local rosetta_json will not be downloaded again save_path is used as the path to save annotation files. If the path has a frame, the data in the frame is obtained. Otherwise, the project_id,, and other values are required.

Return type:

None

Args:

frame:

Frame object

save_path:

Save directory

peoject_id:

rosetta project ID

pool_lst:

pool list

Returns:

Outputs:

path

-{project_id}

--jsons

{task_id}_{frame_id}.json

--images

–{camera_id}

{task_id}_{frame_id).jpg

--pcds

{task_id}_{frame_id}.pcd

--poses

{task_id}_{frame_id).txt

--calibrations

{camera_id}.json