dataloop_download_class #dataloop

dataloop_download_class.py

import dtlpy as dl
import numpy as np
import cv2

import matplotlib
matplotlib.use('TkAgg')
import matplotlib.pyplot as plt


dl.login_m2m(email='', password='')

# == Environment Pointing
dl.add_environment(
    environment='',
    verify_ssl=True,
    alias='',
    gate_url="",
    url=""
)
dl.setenv("")

# == get project by name
project = dl.projects.get(project_name='_Biology')

# == get dataset within project
dataset = project.datasets.get(dataset_name='_Track')

# == ***get all items from dataset***
pages_all = dataset.items.list()

# == FILTER dataset (***get individual annotation***)
'''filters = dl.Filters()
filters.resource = dl.FiltersResource.ANNOTATION
filters.sort = []
pages = dataset.annotations.list(filters=filters)'''

# == FILTER dataset (***get images that are annotated***)
filter = dl.Filters()
filter.add(field='annotated', values=True)
pages_filtered = dataset.items.list(filters=filter)

# == find ids of items that are annotated
file_id_annot = []
for file_annot in pages_filtered.all():
    file_id_annot.append(file_annot.id)

# == find ids of all items
file_id = []
for file in pages_all.all():
    file_id.append(file.id)

# == find ids that are different between two lists which means the ones that are not annotated
file_id_not_annot = set(file_id).difference(set(file_id_annot)).union(set(file_id_annot).difference(set(file_id)))

# == get all items from dataset
#### for file in dataset.items.get_all_items():

for item_id in file_id_annot:

    file = dataset.items.get(item_id=item_id)

    file.download(local_path=r'/Train_Dataset/Dataloop',  # Changed folder name to reflect BBX
                  annotation_options=[dl.VIEW_ANNOTATION_OPTIONS_JSON, 
                                      ])