def generate_set_vals(X_data, Y_data, _rev_data, pr = True):
print("---------------------------------------- Data Set ----------------------------------------")
print('Model: ' + model_name)
prediction = model.predict(X_data)
total = 0
right = 0
for i in range(len(X_data)):
predict = prediction[i][0]
difference = abs(predict - Y_data[i])
predict_stars = dm.to_stars(predict)
actual_stars = dm.to_stars(Y_data[i])
text = str(i + 1) + ') ' + _rev_data[i]
if difference <= allowed_difference:
if pr:
print(dm.truncate_text(text, char_to_display) + ' ----- predicted ' + predict_stars + '----- actually ' + actual_stars)
right += 1
else:
if pr:
print(dm.truncate_text(text, char_to_display) + ' ----- predicted ' + predict_stars + '----- actually ' + actual_stars + ' *****')
total += 1
print('Accuracy:', str(right / total * 100) + '% |', str(right) + '/' + str(total))
def generate_naive(forever = False):
inp = ''
print('Model: ' + model_name)
while forever or not any(inp == w for w in stop_words):
inp = input('Sample review:\n')
prediction = naive_predict(inp)
print(dm.truncate_text('', char_to_display) + ' ----- predicted ' + prediction)
def generate_set_naive(X_data):
print("---------------------------------------- Data Set ----------------------------------------")
print('Model: ' + model_name)
prediction = naive_predict_data(X_data)
for i in range(len(X_data)):
predict = prediction[i]
predict_stars = dm.to_stars(predict)
text = str(i + 1) + ') ' + X_data[i]
print(dm.truncate_text(text, char_to_display) + ' ----- predicted ' + predict_stars)
def generate(forever = False):
inp = ''
print('Model: ' + model_name)
while forever or not any(inp == w for w in stop_words):
inp = input('Sample review:\n')
z = dm.data_to_input(inp)
prediction = model.predict(z)
predict = prediction[0][0]
predict_stars = dm.to_stars(predict)
print(dm.truncate_text('', char_to_display) + ' ----- predicted ' + predict_stars)
def generate_set(X_data, _rev_data):
print("---------------------------------------- Data Set ----------------------------------------")
print('Model: ' + model_name)
z = []
for data in X_data:
z.append(dm.data_to_input(data)[0])
X_data = np.array(z)
prediction = model.predict(X_data)
for i in range(len(_rev_data)):
predict = prediction[i][0]
predict_stars = dm.to_stars(predict)
text = str(i + 1) + ') ' + _rev_data[i]
print(dm.truncate_text(text, char_to_display) + ' ----- predicted ' + predict_stars)
allowed_difference = 0.1
total = 0
right = 0
for i in range(len(X_train)):
predict = prediction[i][0]
difference = abs(predict - Y_train[i])
predict_stars = dm.to_stars(predict)
actual_stars = dm.to_stars(Y_train[i])
text = str(i + 1) + ') ' + _rev_train[i]
if difference <= allowed_difference:
print(dm.truncate_text(text, char_to_display) + ' ----- predicted ' + predict_stars + '----- actually ' + actual_stars)
right += 1
else:
print(dm.truncate_text(text, char_to_display) + ' ----- predicted ' + predict_stars + '----- actually ' + actual_stars + ' *****')
total += 1
print('Loss:', loss, 'Accuracy:', str(right / total * 100) + '% |', str(right) + '/' + str(total))
print("---------------------------------------- TEST SET ----------------------------------------")
loss, acc = model.evaluate(X_test, Y_test, verbose = 0)
prediction = model.predict(X_test)
total = 0
right = 0
total = 0
right = 0
for i in range(len(X_train)):
predict = prediction[i][0]
difference = abs(predict - Y_train[i])
predict_stars = dm.to_stars(predict)
actual_stars = dm.to_stars(Y_train[i])
text = str(i + 1) + ') ' + _rev_train[i]
if difference <= allowed_difference:
print(
dm.truncate_text(text, char_to_display) + ' ----- predicted ' +
predict_stars + '----- actually ' + actual_stars)
right += 1
else:
print(
dm.truncate_text(text, char_to_display) + ' ----- predicted ' +
predict_stars + '----- actually ' + actual_stars + ' *****')
total += 1
print('Loss:', loss, 'Accuracy:',
str(right / total * 100) + '% |',
str(right) + '/' + str(total))
print(
"---------------------------------------- TEST SET ----------------------------------------"
)