def aggregate(project_id, token, href, metadata, environment):
project = AggregationAPI(project_id, environment=environment)
project.__aggregate__()
tarpath = project.__csv_output__(compress=True)
response = send_uploading(metadata, token, href)
url = response.json()["media"][0]["src"]
with open(tarpath, 'rb') as tarball:
requests.put(url, headers={'Content-Type': 'application/x-gzip'}, data=tarball)
os.remove(tarpath)
send_finished(metadata, token, href)
def aggregate(project_id, token, href, metadata, environment):
project = AggregationAPI(project_id, environment=environment)
project.__aggregate__()
tarpath = project.__csv_output__(compress=True)
response = send_uploading(metadata, token, href)
url = response.json()["media"][0]["src"]
with open(tarpath, 'rb') as tarball:
requests.put(url,
headers={'Content-Type': 'application/x-gzip'},
data=tarball)
os.remove(tarpath)
send_finished(metadata, token, href)
#!/usr/bin/env python
import sys
sys.path.append("/home/greg/github/reduction/engine")
sys.path.append("/home/ggdhines/PycharmProjects/reduction/engine")
__author__ = 'greg'
from aggregation_api import AggregationAPI
import numpy
workflow_id = 6
wildebeest = AggregationAPI(6)
aggregations = wildebeest.__aggregate__(workflows = [6],store_values=False)
marking_task = wildebeest.workflows[workflow_id][1].keys()[0]
tools = wildebeest.workflows[workflow_id][1][marking_task]
workflows,versions,instructions,updated_at_timestamps = wildebeest.__get_workflow_details__(workflow_id)
tools_labels = instructions[workflow_id][marking_task]["tools"]
for j,subject_id in enumerate(aggregations):
overall_votes = {int(t_index): [] for t_index in range(len(tools))}
for annotation in wildebeest.__get_raw_classifications__(subject_id,workflow_id):
tool_votes = {int(t_index): 0 for t_index in range(len(tools))}
for task in annotation:
if task["task"] == marking_task:
for marking in task["value"]:
tool_votes[int(marking["tool"])] += 1
for t_index in tool_votes:
overall_votes[t_index].append(tool_votes[t_index])