def vote():
annotator = utils.get_current_annotator()
if annotator is not None:
if annotator.prev.id == int(
request.form['prev_id']) and annotator.next.id == int(
request.form['next_id']):
if request.form['action'] == 'Skip':
annotator.ignore.append(annotator.next)
else:
if request.form['action'] == 'Previous':
utils.perform_vote(annotator, next_won=False)
decision = Decision(annotator,
winner=annotator.prev,
loser=annotator.next)
elif request.form['action'] == 'Current':
utils.perform_vote(annotator, next_won=True)
decision = Decision(annotator,
winner=annotator.next,
loser=annotator.prev)
db.session.add(decision)
annotator.next.viewed.append(annotator)
annotator.prev = annotator.next
annotator.ignore.append(annotator.prev)
annotator.next = utils.choose_next(annotator)
db.session.commit()
return redirect(url_for('index'))
def create_decision(request, status, template_name):
decision = Decision(status=status)
decision.author = request.user
if request.method == "POST":
return _process_post_and_redirect(request, decision, template_name)
form = DecisionForm(instance=decision)
data = dict(form=form,tab=status)
context = RequestContext(request, data)
return render_to_response(template_name, context)
def main():
args = get_arguments()
print(args)
MODEL_DIR = os.path.join(ROOT_DIR, "logs/decision")
data_dir = "./data/"
config = CityscapeConfig()
config.FLOW = args.flowmodel
decision = Decision(config, dropout=0.5)
decision.train(data_dir, MODEL_DIR, epochs=200)
def modify_decision(request, decision_id = None, status_id = None):
if decision_id is None:
decision = Decision()
if status_id is not None:
decision.status = int(status_id)
else:
decision = get_object_or_404(Decision, id = decision_id)
if request.method == "POST":
if request.POST.get('submit', None) == "Cancel":
return_page = unicode(decision.status_text())
return HttpResponseRedirect(reverse(listing, args=[return_page]))
else:
decision_form = DecisionForm(data=request.POST,
instance=decision)
feedback_formset = FeedbackFormSet(data=request.POST,
instance=decision)
if decision_form.is_valid():
decision = decision_form.save(commit=False)
feedback_formset = FeedbackFormSet(request.POST,
instance=decision)
if feedback_formset.is_valid():
decision.save(request.user)
if decision_form.cleaned_data['watch']:
decision.add_watcher(request.user)
else:
decision.remove_watcher(request.user)
feedback_formset.save()
return_page = unicode(decision.status_text())
return HttpResponseRedirect(reverse(listing, args=[return_page]))
else:
feedback_formset = FeedbackFormSet(instance=decision)
decision_form = DecisionForm(instance=decision)
return render_to_response('decision_edit.html',
RequestContext(request,
dict(decision_form=decision_form, feedback_formset=feedback_formset)))
def _get_test_contest_decisions(contest):
random.seed()
decisions = []
ballot_ids = [str(uuid.uuid4()) for x in range(7)]
num_decisions = random.randint(90, 200)
for i in range(num_decisions):
voter_opinions = {}
write_in_names = []
if contest.decision_type == contest.DECISION_TYPE_VOTE_YES_NO:
voter_opinions[0] = random.randint(0, 1)
else:
if contest.allow_write_ins and (random.randint(1, 10000) %
25) == 0:
# # enter a write in
write_in_names.append("Write In {0}".format(
random.randint(1, 4)))
voter_opinions[len(contest.contestants)] = 1
else:
# #pick a candidate ar random
voter_opinions[random.randint(0,
len(contest.contestants) -
1)] = 1
decision = Decision(
contest, {
'decision_id': str(uuid.uuid4()),
'contest_id': contest.id,
'ballot_id': ballot_ids[random.randint(1, 10000) % 7],
'timestamp': datetime.now(),
'contest': contest,
'authoritative': True,
'latest': True,
'voter_id': str(uuid.uuid4()),
'voter_opinions': voter_opinions,
'write_in_names': write_in_names
})
decisions.append(decision)
return decisions
config.BACKBONE = args.resnetmodel
config.IMAGE_SHAPE = [1024, 1024, 6]
config.POST_NMS_ROIS_INFERENCE = 500
#config.display()
# Validation dataset
dataset = CityscapeDataset()
dataset.load_cityscape(args.data_dir, "val", args.num_frames, args.fix)
dataset.prepare()
print("Image Count: {}".format(len(dataset.image_ids)))
resnet = ResNet(config=config)
flownet = FlowNet(config=config)
maskrcnn = MaskRCNN(config=config)
warp = Warp(config=config)
decision = Decision(config=config)
model_path = args.restore_from
resnet.load_weights(model_path, by_name=True)
flownet.load_weights(model_path, by_name=True)
maskrcnn.load_weights(model_path, by_name=True)
decision.load_weights(args.decision_from, by_name=True)
AP50s = []
APs = []
seg_step = 0
flow_step = 0
score = 0
if args.method == 2:
target = -args.target
else:
def main():
args = get_arguments()
print(args)
config = CityscapeConfig()
config.FLOW = args.flowmodel
config.BACKBONE = args.resnetmodel
config.IMAGE_SHAPE = [1024, 1024, 6]
config.POST_NMS_ROIS_INFERENCE = 500
#config.display()
resnet = ResNet(config=config)
flownet = FlowNet(config=config)
maskrcnn = MaskRCNN(config=config)
warp = Warp(config=config)
decision = Decision(config=config)
model_path = args.restore_from
resnet.load_weights(model_path, by_name=True)
flownet.load_weights(model_path, by_name=True)
maskrcnn.load_weights(model_path, by_name=True)
decision.load_weights(args.decision_from, by_name=True)
seg_step = 0
flow_step = 0
target = args.target
list_file = open(args.data_list, 'r')
if args.is_save and not os.path.exists(args.save_dir):
os.makedirs(args.save_dir)
for step in range(args.num_steps):
f1 = list_file.readline().split('\n')[0]
f1 = os.path.join(args.data_dir, f1)
current = np.expand_dims(skimage.io.imread(f1), 0)
current, image_metas, window = mold_inputs(config, current)
image = current[0] + config.MEAN_PIXEL
if step == 0:
seg_step += 1
key_P2, key_P3, key_P4, key_P5, key_P6 = resnet.keras_model.predict(
current)
key = current
P2, P3, P4, P5, P6 = key_P2, key_P3, key_P4, key_P5, key_P6
else:
images = np.concatenate([current, key], 3)
flow, flow_feature = flownet.keras_model.predict(images)
score = decision.keras_model.predict(flow_feature)[0][0]
print("step: {:4d} predict score: {:.3f} target: {:.2f}".format(
step, score, target))
if score < target:
if args.dynamic and target < 50:
target -= 0.5
seg_step += 1
key_P2, key_P3, key_P4, key_P5, key_P6 = resnet.keras_model.predict(
current)
key = current
P2, P3, P4, P5, P6 = key_P2, key_P3, key_P4, key_P5, key_P6
else:
if args.dynamic and target < 95:
target += 0.25
flow_step += 1
P2, P3, P4, P5, P6 = warp.predict(
[key_P2, key_P3, key_P4, key_P5, key_P6, flow])
inputs = [image_metas, P2, P3, P4, P5, P6]
result = maskrcnn.detect_molded(inputs)
# Save
if args.is_save:
save_name = args.save_dir + 'mask' + str(step) + '.png'
colors = np.array(label_colors) / 255.0
pred_img = visualize.display_instances(image,
result['rois'],
result['masks'],
result['class_ids'],
class_names,
result['scores'],
colors=colors,
save_name=save_name)
print("segmentation steps:", seg_step, "flow steps:", flow_step)
def get_decision(self):
return Decision(datetime.utcnow(), self.__decision, self.name)
def modify_decision(request, decision_id=None, status_id=None):
if decision_id is None:
decision = Decision()
if status_id is not None:
decision.status = int(status_id)
else:
decision = get_object_or_404(Decision, id=decision_id)
if request.method == "POST":
if request.POST.get('submit', None) == "Cancel":
return_page = unicode(decision.status_text())
return HttpResponseRedirect(reverse(listing, args=[return_page]))
else:
decision_form = DecisionForm(data=request.POST, instance=decision)
feedback_formset = FeedbackFormSet(data=request.POST,
instance=decision)
if decision_form.is_valid():
decision = decision_form.save(commit=False)
feedback_formset = FeedbackFormSet(request.POST,
instance=decision)
if feedback_formset.is_valid():
decision.save(request.user)
if decision_form.cleaned_data['watch']:
decision.add_watcher(request.user)
else:
decision.remove_watcher(request.user)
feedback_formset.save()
return_page = unicode(decision.status_text())
return HttpResponseRedirect(
reverse(listing, args=[return_page]))
else:
feedback_formset = FeedbackFormSet(instance=decision)
decision_form = DecisionForm(instance=decision)
return render_to_response(
'decision_edit.html',
RequestContext(
request,
dict(decision_form=decision_form,
feedback_formset=feedback_formset)))
def _get_real_contest_decisions(contest):
decisions = []
results = api.ballot_get_decisions_by_contest(contest.id)
if results:
decisions = [Decision(contest, d) for d in results]
return decisions