def submit_text(request):
if request.method == 'POST':
try:
text = request.POST['input_text']
filename_names = os.path.join(CLASSIFIER_ROOT, "names.pkl")
with open(filename_names, "rb") as f:
names_loaded = pickle.load(f)
text_classifier = TextClassifier()
text_classifier.load(os.path.join(CLASSIFIER_ROOT, "class.pkl"))
predicted = text_classifier.predict([text])
offset = 0
mask = text_classifier.get_support()
tags = []
for i in range(len(predicted[0])):
while not mask[offset]:
offset += 1
if predicted[0][i] != 0:
tags.append(names_loaded[offset])
offset += 1
return HttpResponse(', '.join(tags))
except Exception as e:
return HttpResponse(e.message)
def test_advanced_quality(self):
sgdc = SGDClassifier(alpha=0.0001, class_weight=None, epsilon=0.1,
eta0=0.0, fit_intercept=True, l1_ratio=0.15,
learning_rate='optimal', loss='log', n_iter=5, n_jobs=1,
penalty='l2', power_t=0.5, random_state=None, shuffle=True,
verbose=0, warm_start=False)
textClassifier = TextClassifier(base_classifiers =[sgdc, LinearSVC()]) #, multilabel = False)
# textClassifier.fit(self.X_train, self.Y_train)
textClassifier.fit(self.X_train, self.Y_train, multilabel=False)
predicted = textClassifier.predict(self.X_test)
advanced_quality = 0.8
self.assertGreaterEqual(metrics.precision_score(self.Y_test, predicted, average = 'micro'), advanced_quality)
