First steps with Remo python library¶
Create and visualize a dataset¶
Let's create a new dataset and upload some annotations.
import remo
import pandas as pd
(\(\
(>':') Remo server is running: {'app': 'remo', 'version': '0.3.5-455-g887e55a2'}
urls = ['https://remo-scripts.s3-eu-west-1.amazonaws.com/open_images_sample_dataset.zip']
remo.create_dataset(name = 'open images detection',
urls = urls,
annotation_task = "Object detection")
{'files uploaded': 10, 'annotations': 10, 'errors': []}
Dataset 14 - 'open images detection'
You can read more about what type of annotation tasks and formats we support in our documentation.
We can easily list all datasets and retrieve one
remo.list_datasets()
[Dataset 2 - 'OCR_symbols',
Dataset 6 - 'cars_detection',
Dataset 14 - 'open images detection']
# make sure to use the right ID when running the tutorial
my_dataset = remo.get_dataset(14)
my_dataset.view()
Open http://localhost:8123/datasets/14
Visualize Annotation Statistics¶
To explore annotations, we can print the stats of the annotation sets or open the interactive UI
my_dataset.get_annotation_statistics()
[{'AnnotationSet ID': 10,
'AnnotationSet name': 'Object detection',
'n_images': 10,
'n_classes': 18,
'n_objects': 98,
'top_3_classes': [{'name': 'Fruit', 'count': 27},
{'name': 'Sports equipment', 'count': 12},
{'name': 'Human arm', 'count': 10}],
'creation_date': None,
'last_modified_date': '2020-02-23T20:55:51.040660Z'}]
my_dataset.view_annotation_stats()
Open http://localhost:8123/annotation-detail/10/intro
Export Annotations¶
We can easily export annotations in a standardised format, and use them for training a model or further analysis
my_dataset.export_annotations_to_file('output.csv', annotation_format='csv')
Further SDK functionalities¶
Refer to the other tutorials and the documentation to explore further the SDK.
Other functionalities include:
- Manipulating annotation sets from code
- Custom uploading of annotations, predictions and images
- Advanced images search
- Organising data in virtual folders