mirror of https://github.com/doccano/doccano.git
pythondatasetsactive-learningtext-annotationdatasetnatural-language-processingdata-labelingmachine-learningannotation-tool
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
149 lines
6.0 KiB
149 lines
6.0 KiB
import pathlib
|
|
from unittest.mock import patch
|
|
|
|
from auto_labeling_pipeline.mappings import AmazonComprehendSentimentTemplate
|
|
from auto_labeling_pipeline.models import RequestModelFactory
|
|
from model_mommy import mommy
|
|
from rest_framework import status
|
|
from rest_framework.reverse import reverse
|
|
|
|
from api.models import DOCUMENT_CLASSIFICATION, IMAGE_CLASSIFICATION
|
|
from api.tests.api.utils import (CRUDMixin, make_auto_labeling_config, make_doc, make_image,
|
|
prepare_project)
|
|
|
|
data_dir = pathlib.Path(__file__).parent / 'data'
|
|
|
|
|
|
class TestConfigParameter(CRUDMixin):
|
|
|
|
def setUp(self):
|
|
self.project = prepare_project(task=DOCUMENT_CLASSIFICATION)
|
|
self.data = {
|
|
'model_name': 'GCP Entity Analysis',
|
|
'model_attrs': {'key': 'hoge', 'type': 'PLAIN_TEXT', 'language': 'en'},
|
|
'text': 'example'
|
|
}
|
|
self.url = reverse(viewname='auto_labeling_parameter_testing', args=[self.project.item.id])
|
|
|
|
@patch('auto_labeling.views.AutoLabelingConfigParameterTest.send_request', return_value={})
|
|
def test_called_with_proper_model(self, mock):
|
|
self.assert_create(self.project.users[0], status.HTTP_200_OK)
|
|
_, kwargs = mock.call_args
|
|
expected = RequestModelFactory.create(self.data['model_name'], self.data['model_attrs'])
|
|
self.assertEqual(kwargs['model'], expected)
|
|
|
|
@patch('auto_labeling.views.AutoLabelingConfigParameterTest.send_request', return_value={})
|
|
def test_called_with_text(self, mock):
|
|
self.assert_create(self.project.users[0], status.HTTP_200_OK)
|
|
_, kwargs = mock.call_args
|
|
self.assertEqual(kwargs['example'], self.data['text'])
|
|
|
|
@patch('auto_labeling.views.AutoLabelingConfigParameterTest.send_request', return_value={})
|
|
def test_called_with_image(self, mock):
|
|
self.data['text'] = str(data_dir / 'images/1500x500.jpeg')
|
|
self.assert_create(self.project.users[0], status.HTTP_200_OK)
|
|
_, kwargs = mock.call_args
|
|
self.assertEqual(kwargs['example'], self.data['text'])
|
|
|
|
|
|
class TestTemplateMapping(CRUDMixin):
|
|
|
|
def setUp(self):
|
|
self.project = prepare_project(task=DOCUMENT_CLASSIFICATION)
|
|
self.data = {
|
|
'response': {
|
|
'Sentiment': 'NEUTRAL',
|
|
'SentimentScore': {
|
|
'Positive': 0.004438233096152544,
|
|
'Negative': 0.0005306027014739811,
|
|
'Neutral': 0.9950305223464966,
|
|
'Mixed': 5.80838445785048e-7
|
|
}
|
|
},
|
|
'template': AmazonComprehendSentimentTemplate().load()
|
|
}
|
|
self.url = reverse(viewname='auto_labeling_template_test', args=[self.project.item.id])
|
|
|
|
def test_template_mapping(self):
|
|
response = self.assert_create(self.project.users[0], status.HTTP_200_OK)
|
|
expected = [{'label': 'NEUTRAL'}]
|
|
self.assertEqual(response.json(), expected)
|
|
|
|
def test_json_decode_error(self):
|
|
self.data['template'] = ''
|
|
self.assert_create(self.project.users[0], status.HTTP_400_BAD_REQUEST)
|
|
|
|
|
|
class TestLabelMapping(CRUDMixin):
|
|
|
|
def setUp(self):
|
|
self.project = prepare_project(task=DOCUMENT_CLASSIFICATION)
|
|
self.data = {
|
|
'response': [{'label': 'NEGATIVE'}],
|
|
'label_mapping': {'NEGATIVE': 'Negative'}
|
|
}
|
|
self.url = reverse(viewname='auto_labeling_mapping_test', args=[self.project.item.id])
|
|
|
|
def test_label_mapping(self):
|
|
response = self.assert_create(self.project.users[0], status.HTTP_200_OK)
|
|
expected = [{'label': 'Negative'}]
|
|
self.assertEqual(response.json(), expected)
|
|
|
|
|
|
class TestConfigCreation(CRUDMixin):
|
|
|
|
def setUp(self):
|
|
self.project = prepare_project(task=DOCUMENT_CLASSIFICATION)
|
|
self.data = {
|
|
'model_name': 'Amazon Comprehend Sentiment Analysis',
|
|
'model_attrs': {
|
|
'aws_access_key': 'str',
|
|
'aws_secret_access_key': 'str',
|
|
'region_name': 'us-east-1',
|
|
'language_code': 'en'
|
|
},
|
|
'template': AmazonComprehendSentimentTemplate().load(),
|
|
'label_mapping': {'NEGATIVE': 'Negative'}
|
|
}
|
|
self.url = reverse(viewname='auto_labeling_configs', args=[self.project.item.id])
|
|
|
|
def test_create_config(self):
|
|
response = self.assert_create(self.project.users[0], status.HTTP_201_CREATED)
|
|
self.assertEqual(response.data['model_name'], self.data['model_name'])
|
|
|
|
def test_list_config(self):
|
|
mommy.make('AutoLabelingConfig', project=self.project.item)
|
|
response = self.assert_fetch(self.project.users[0], status.HTTP_200_OK)
|
|
self.assertEqual(len(response.data), 1)
|
|
|
|
|
|
class TestAutoLabelingText(CRUDMixin):
|
|
|
|
def setUp(self):
|
|
self.project = prepare_project(task=DOCUMENT_CLASSIFICATION)
|
|
make_auto_labeling_config(self.project.item)
|
|
self.example = make_doc(self.project.item)
|
|
self.url = reverse(viewname='auto_labeling_annotation', args=[self.project.item.id, self.example.id])
|
|
|
|
@patch('auto_labeling.views.execute_pipeline', return_value=[])
|
|
def test_text_task(self, mock):
|
|
self.assert_create(self.project.users[0], status.HTTP_201_CREATED)
|
|
_, kwargs = mock.call_args
|
|
self.assertEqual(kwargs['text'], self.example.text)
|
|
|
|
|
|
class TestAutoLabelingImage(CRUDMixin):
|
|
|
|
def setUp(self):
|
|
self.project = prepare_project(task=IMAGE_CLASSIFICATION)
|
|
make_auto_labeling_config(self.project.item)
|
|
filepath = data_dir / 'images/1500x500.jpeg'
|
|
self.example = make_image(self.project.item, str(filepath))
|
|
self.url = reverse(viewname='auto_labeling_annotation', args=[self.project.item.id, self.example.id])
|
|
|
|
@patch('auto_labeling.views.execute_pipeline', return_value=[])
|
|
def test_text_task(self, mock):
|
|
self.assert_create(self.project.users[0], status.HTTP_201_CREATED)
|
|
_, kwargs = mock.call_args
|
|
expected = str(self.example.filename)
|
|
self.assertEqual(kwargs['text'], expected)
|