train_dir = 'Data/train'
test_dir = 'Data/test'
validation_dir = 'Data/validation'
train_datagen = ImageDataGenerator(rescale=1/255,
                                   rotation_range=10,
                                   width_shift_range=0.2,  #圖片水平偏移的角度
                                   height_shift_range=0.2,  #圖片數(shù)值偏移的角度
                                   shear_range=0.2,  #剪切強度 
                                   zoom_range=0.2,   #隨機縮放的幅度
                                   horizontal_flip=True,   #是否進行隨機水平翻轉(zhuǎn)
#                                    fill_mode='nearest'
                                  )
train_generator = train_datagen.flow_from_directory(train_dir,
                 (224,224),batch_size=1,class_mode='binary',shuffle=False)
test_datagen = ImageDataGenerator(rescale=1/255)
test_generator = test_datagen.flow_from_directory(test_dir,
                 (224,224),batch_size=1,class_mode='binary',shuffle=True)
validation_datagen = ImageDataGenerator(rescale=1/255)
validation_generator = validation_datagen.flow_from_directory(
                validation_dir,(224,224),batch_size=1,class_mode='binary')
print(train_datagen)
print(test_datagen)
print(train_datagen)

train_data=[]
train_labels=[]
a=0
for data_train, labels_train in train_generator:
    train_data.append(data_train)
    train_labels.append(labels_train)
    a=a+1
    if a>1999:
        break
x_train=np.array(train_data)
y_train=np.array(train_labels)
x_train=x_train.reshape(2000,224,224,3)

test_data=[]
test_labels=[]
a=0
for data_test, labels_test in test_generator:
    test_data.append(data_test)
    test_labels.append(labels_test)
    a=a+1
    if a>999:
        break
x_test=np.array(test_data)
y_test=np.array(test_labels)
x_test=x_test.reshape(1000,224,224,3)

validation_data=[]
validation_labels=[]
a=0
for data_validation, labels_validation in validation_generator:
    validation_data.append(data_validation)
    validation_labels.append(labels_validation)
    a=a+1
    if a>999:
        break
x_validation=np.array(validation_data)
y_validation=np.array(validation_labels)
x_validation=x_validation.reshape(1000,224,224,3)

model1 = tf.keras.models.Sequential([
    # 第一層卷積，卷積核為，共16個，輸入為150*150*1
    tf.keras.layers.Conv2D(16,(3,3),activation='relu',padding='same',input_shape=(224,224,3)),
    tf.keras.layers.MaxPooling2D((2,2)),
    
    # 第二層卷積，卷積核為3*3，共32個，
    tf.keras.layers.Conv2D(32,(3,3),activation='relu',padding='same'),
    tf.keras.layers.MaxPooling2D((2,2)),
    
    # 第三層卷積，卷積核為3*3，共64個，
    tf.keras.layers.Conv2D(64,(3,3),activation='relu',padding='same'),
    tf.keras.layers.MaxPooling2D((2,2)),
    
    # 數(shù)據(jù)鋪平
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(64,activation='relu'),
    tf.keras.layers.Dropout(0.5),
    tf.keras.layers.Dense(1,activation='sigmoid')
])
print(model1.summary())

model1.compile(optimize=tf.keras.optimizers.SGD(0.00001),
             loss=tf.keras.losses.binary_crossentropy,
             metrics=['acc'])
history1=model1.fit(x_train,y_train,
# 					validation_split=(0~1)   選擇一定的比例用于驗證集，可被validation_data覆蓋
                  validation_data=(x_validation,y_validation),
                  batch_size=10,
                  shuffle=True,
                  epochs=10)
model1.save('cats_and_dogs_plain1.h5')
print(history1)

plt.plot(history1.epoch,history1.history.get('acc'),label='acc')
plt.plot(history1.epoch,history1.history.get('val_acc'),label='val_acc')
plt.title('正確率')
plt.legend()

model1.evaluate(x_validation,y_validation)