classifier.add(Dense(units = 128, activation = 'relu'))
classifier.add(Dense(units = 4, activation = 'softmax'))

# Compiling the CNN
 classifier.compile(optimizer = 'adam', loss =
'categorical_crossentropy', metrics = ['accuracy'])

# Part 2 - Fitting the CNN to the images
from keras.preprocessing.image import ImageDataGenerator
train_datagen = ImageDataGenerator(rescale = 1./255,
shear_range = 0.2,
zoom_range = 0.2,
horizontal_flip = True)
test_datagen = ImageDataGenerator(rescale = 1./255)
training_set = train_datagen.flow_from_directory('dataset/dota_training_set',
target_size = (64, 64),
batch_size = 32,
class_mode = 'categorical')
test_set = test_datagen.flow_from_directory('dataset/dota_test_set',
target_size = (64, 64),
batch_size = 32,
class_mode = 'categorical')
classifier.fit_generator(training_set,
steps_per_epoch = 1000,
epochs = 3,
validation_data = test_set,
validation_steps = 500)

# Part 3 - Making new predictions
import numpy as np
from keras.preprocessing import image
test_image = image.load_img('dataset/single_prediction/test_2.jpg', target_size = (64, 64))
test_image = image.img_to_array(test_image)
test_image = np.expand_dims(test_image, axis = 0)
result = classifier.predict(test_image)
training_set.class_indices
if result[0][0] == 1.0:
    prediction = 'sven'
    print("This is a Sven.")
if result[0][0] == 2.0:
    prediction = 'mirana'
    print("This is a mirana.")
if result[0][0] == 3.0:
    prediction = 'Wraith_King'
    print("This is a Wraith King.")
if result[0][0] == 4.0:
    prediction = 'Phantom_Lancer'
    print("This is a Phantom Lancer.")
else:
    prediction = 'non_dota hero'
    print("This is a non dota hero.")