def load_classifier():
if not os.path.exists('classifier.db'):
classifier = train_all(None , corpus)
classifier.save('classifier.db')
return Classifier.load('classifier.db')
def load_classifier():
if not os.path.exists('classifier.db'):
sys.stderr.write('Unable to load classifier.db -- please run this script first')
sys.exit(1)
return Classifier.load('classifier.db')
def load_classifier():
if not os.path.exists('classifier.db'):
sys.stderr.write('Unable to load classifier.db -- please run this script first')
sys.exit(1)
return Classifier.load('classifier.db')
subj = ""
for line in text.splitlines():
if line.startswith("subject:"):
is_subj = True
subj = line[8:]
else:
lines.append(line)
return email_extract(subj, " ".join(lines), min_len, max_len)
def test_enron_files(classifier, path, label, selector=None):
files = get_file_list(path, selector)
correct = total = 0
for filename in files:
with open(os.path.join(path, filename)) as fh:
contents = fh.read()
features = enron_email_extract(contents)
res = classifier.classify(features)
best = res[0][0]
if best == label:
correct += 1
total += 1
pct = 100 * (float(correct) / total)
print 'Accuracy of "%s": %s%% based on %s documents' % (label, pct, total)
if __name__ == "__main__":
classifier = Classifier.load("classifier.bin")
for d, l in get_dir_and_labels():
test_enron_files(classifier, d, l)
