model = models.Sequential()
model.add(layers.Conv2D(32, (3, 3), activation='relu',
                    input_shape=(1500, 1500, 3)))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(128, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(256, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(256, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Flatten())
model.add(layers.Dense(512, activation='relu'))
model.add(layers.Dense(1, activation='sigmoid'))

# All images will be rescaled by 1./255
train_datagen = ImageDataGenerator(rescale=1./255)
test_datagen = ImageDataGenerator(rescale=1./255)

train_generator = train_datagen.flow_from_directory(
    # This is the target directory
    train_dir,
    # All images will be resized to 1500x1500
    target_size=(1500, 1500),
    batch_size=2,
    # Since we use binary_crossentropy loss, we need binary labels
    class_mode='binary')

 validation_generator = test_datagen.flow_from_directory(
    validation_dir,
    target_size=(1500, 1500),
    batch_size=2,
    class_mode='binary')

test_generator = test_datagen.flow_from_directory(
    test_dir,
    target_size=(1500, 1500),
    batch_size=2,
    class_mode='binary')

  probabilities = model.predict_generator(test_generator,5)

array([[2.6628117e-28],
   [6.6314442e-06],
   [3.2372427e-20],
   [7.8302348e-04],
   [1.0000000e+00],
   [1.0000000e+00],
   [1.0000000e+00],
   [8.4590050e-14],
   [9.9938679e-01],
   [3.6370282e-25]], dtype=float32)