annolab 0.3.0

Official SDK for the AnnoLab Platform

This is the official python SDK for AnnoLab, the ML platform for NLP projects.

AnnoLab Website

Getting Started

Assuming that you have Python and virtualenv installed, set up your environment and install the required dependencies like this or you can install the library using pip:

$ virtualenv venv
$ . venv/bin/activate
$ python -m pip install annolab

Using the AnnoLab SDK

To get started, ensure you have an annolab account at https://app.annolab.ai/signup and have created an API Key. Instructions for creating an API Key may be found at https://docs.annolab.ai/.

Configure the sdk with your api key using one of the following two methods.

1. Create an instance of the SDK passing your api_key.

>>> from annolab import Annolab
>>> lab = AnnoLab(api_key='YOUR_API_KEY')

2. Or set a global api key. All subsequent uses of the sdk will use this global key for authentication.

>>> import annolab
>>> from annolab import Annolab
>>>
>>> annolab.api_key = 'YOUR_API_KEY'
>>> lab = AnnoLab()

Usage Examples

Creating a project.

lab.create_project('My New Project')
# OR
lab.create_project(name='My New Project', owner_name='AnnoLab')

Getting an existing project.

lab.find_project('My New Project')
# OR
lab.find_project(name='My New Project', owner_name='AnnoLab')

Creating a new text source. Will be added to the “Uploads” directory by default.

project = lab.find_project('My New Project')
project.create_text_source(name='New Source', text='Some text or tokens for annotation.')
# Specifying a directory
project.create_text_source(
  name='New Source',
  text='Some text or tokens for annotation.',
  directory='Uploads'
)

Creating a new pdf source from a file. Will be added to the “Uploads” directory by default.

project = annolab.find_project('My New Project')
project.create_pdf_source(file='/path/to/file')
project.create_pdf_source(file='/path/to/file', name='custom_name.pdf', directory='Uploads')

# You may also pass a filelike object or bytes. "name" is required when doing so.
project.create_pdf_source(file=open('myfile.pdf', 'r+b'), name='myfile.pdf')

Creating a new pdf source from a web source.

project = annolab.find_project('My New Project')
project.create_pdf_source_from_web(url='https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf', name='mypdf.pdf')

Adding annotations.

  project.create_annotations(
    source_name='New Source',
    annotations=[
        { 'type': 'one', 'value': 'value one', 'offsets': [0, 10]},
        { 'type': 'two', 'value': 'two', 'offsets': [10, 20] }
    ],
)

Adding annotations with relations.

  project.create_annotations(
    source_name='New Source',
    annotations=[
        { 'clientId': 1, 'type': 'one', 'value': 'value one', 'offsets': [0, 10]},
        { 'clientId': 2, 'type': 'two', 'value': 'two', 'offsets': [10, 20] }
    ],
    relations=[
      { 'annotations': [1, 2] }
    ]
)

Exporting a project.

project.export(filepath='/path/to/outfile.zip')

# With options
project.export(
  filepath='/path/to/outfile.zip',
  source_ids=[1,2,3],
  layers=['GoldSet'],
  include_annotation_types=True,
  include_sources=True
)