def setUp(self):
self.tasks = {}
for lang in [('en', 'E'), ('ja', 'J')]:
queryfilepath = create_query_subset(
'./data/MC2-training/%s/1C2-%s-queries.tsv' % lang,
'./data/MC2-training-documents/1C2-%s.INDX/' % lang[1])
intentfilepath = create_tmp_intent_file(queryfilepath)
self.tasks[lang[1]] = Task.read(queryfilepath,
'./data/MC2-training/%s/1C2-%s-iunits.tsv' % lang,
'./data/MC2-training-documents/1C2-%s.INDX/' % lang[1],
'./data/MC2-training-documents/1C2-%s.HTML/' % lang[1],
intentfilepath)
def test_task_read(self):
"""
Task.read (with intents)
"""
tasks = Task.read(
self.queryfilepath,
"./data/MC2-training/en/1C2-E-iunits.tsv",
"./data/MC2-training-documents/1C2-E.INDX/",
"./data/MC2-training-documents/1C2-E.HTML/",
self.intentfilepath,
)
self.assertEqual(len(tasks), 5)
self.assertEqual(tasks[0].query.qid, "1C2-E-0001")
self.assertEqual(len(tasks[0].intents), 5)
self.assertEqual(tasks[0].intents[0].qid, "1C2-E-0001")
self.assertEqual(tasks[0].intents[0].iid, "1C2-E-0001-INTENT0001")
def test_task_read(self):
'''
Task.read
'''
tasks = Task.read(self.queryfilepath,
'./data/MC2-training/en/1C2-E-iunits.tsv',
'./data/MC2-training-documents/1C2-E.INDX/',
'./data/MC2-training-documents/1C2-E.HTML/')
self.assertEqual(len(tasks), 5)
self.assertEqual(tasks[0].query.qid, '1C2-E-0001')
self.assertEqual(len(tasks[0].iunits), 19)
self.assertEqual(tasks[0].iunits[0].qid, '1C2-E-0001')
self.assertEqual(tasks[0].iunits[0].uid, '1C2-E-0001-0001')
self.assertEqual(len(tasks[0].indices), 213)
self.assertEqual(tasks[0].indices[0].qid, '1C2-E-0001')
self.assertEqual(tasks[0].indices[0].rank, 1)
Index.qid: Query ID
Index.filepath: filepath of an HTML file
Index.rank: rank in a search engine result page
Index.title: webpage title
Index.url: webpage url
Index.body: summary of the webpage
"""
def rank(self, task):
"""
Output ranked pairs of an iUnits and a score
e.g. Random ranking method
return [(i, 0) for i in task.iunits]
"""
return [(i, 0) for i in task.iunits]
if __name__ == "__main__":
from mobileclick import Task
tasks = Task.read(
"data/MC2-training/en/1C2-E-queries.tsv",
"data/MC2-training/en/1C2-E-iunits.tsv",
"data/MC2-training-documents/1C2-E.INDX",
"data/MC2-training-documents/1C2-E.HTML",
)
method = YourRankingMethod()
run = method.generate_run("YourRun", "This is your run", tasks)
run.save("./")
summary.add(iunit6) # added to the first layer
summary.add(iunit7, intent1.iid) # added to the second layer
summary.add(intent3) # added to the first layer
summary.add(iunit8, intent3.iid) # added to the second layer
The resultant summary is
First layer:
iunit1, iunit2, intent1, intent2, iunit6, intent3
Second layer:
intent1: iunit3, iunit4, iunit7
intent2: iunit3, iunit5
intent3: iunit8
e.g. Random summarization method
return Summary(task.query.qid, task.iunits)
'''
return Summary(task.query.qid, task.iunits)
if __name__ == '__main__':
from mobileclick import Task
tasks = Task.read(
"data/MC2-test/en/MC2-E-queries.tsv",
"data/MC2-test/en/MC2-E-iunits.tsv",
"data/MC2-test-documents/MC2-E.INDX",
"data/MC2-test-documents/MC2-E.HTML",
"data/MC2-test/en/MC2-E-intents.tsv")
method = YourSummarizationMethod()
run = method.generate_run("YourRun", "This is your run", tasks)
run.save('./')
