def main(train_dir, labels=['spam', 'ham']):
print('Train dir: %s' % train_dir)
paths = {}
for label in labels:
paths[label] = []
for label, path in read_training_data(train_dir):
if label == None:
print('Label is None for Path=%s' % path)
paths[label].append(path)
model = NaiveBayesModel(paths)
fname = 'nbmodel.txt'
model.save_to_path(fname)
def main(data_dir, model_path, out_path):
print("Reading data from %s\nReading model from %s\n" \
"Storing output to %s" % (data_dir, model_path, out_path))
model = NaiveBayesModel.load_from_path(model_path)
data = read_training_data(data_dir)
with open(out_path, 'w', 1, encoding=ENCODING) as outf:
for label, path in data:
with open(path, 'r', encoding=ENCODING) as inf:
doc = inf.read()
preds = model.predict(doc)
maxlabel = max(preds, key=preds.get)
outf.write("%s %s\n" % (maxlabel, path))
