# -*- coding: utf-8 -*-
import data_helpers
import keras
import sys
from keras.utils import plot_model
reload(sys)
sys.setdefaultencoding('utf-8')
input = sys.argv[1][0:270]
x = data_helpers.filterinput(input)
#ensemble3_1
model = keras.models.load_model('save_ensemble3_1.h5')
#plot_model(model, to_file='model.png', show_shapes=True, show_layer_names=True)
y = model.predict(x)
#resultstr = input + "\n"
resultstr = ""
if round(y) == 1.0:
resultstr += "Not Spam"
else:
resultstr += "Spam"
print "---RESULT---" + resultstr + "---RESULT---"
# -*- coding: utf-8 -*-
import data_helpers
import keras
import sys
from keras.models import Model
reload(sys)
sys.setdefaultencoding('utf-8')
x = data_helpers.filterinput(sys.argv[1])
model = keras.models.load_model('./save_tmp.h5')
newmodel = Model(model.input, model.layers[5].output)
y = model.predict(x)
y2 = newmodel.predict(x)[0][0:len(sys.argv[1].split())]
resultstr = ""
if round(y) == 1.0:
resultstr = "NOT SPAM"
else:
resultstr = "SPAM"
index = -1
biggestScore = -1000
biggestWord = ""
for i in y2:
index += 1
if biggestScore < i:
biggestScore = i
biggestWord = sys.argv[1].split()[index]