# In this exercise you will train a CNN on the FULL Cats-v-dogs dataset
# This will require you doing a lot of data preprocessing because
# the dataset isn't split into training and validation for you
# This code block has all the required inputs
import os
import zipfile
import random
import tensorflow as tf
from tensorflow.keras.optimizers import RMSprop
from tensorflow.keras.preprocessing.image import ImageDataGenerator
from shutil import copyfile
# This code block downloads the full Cats-v-Dogs dataset and stores it as 
# cats-and-dogs.zip. It then unzips it to /tmp
# which will create a tmp/PetImages directory containing subdirectories
# called 'Cat' and 'Dog' (that's how the original researchers structured it)
# If the URL doesn't work, 
# .  visit https://www.microsoft.com/en-us/download/confirmation.aspx?id=54765
# And right click on the 'Download Manually' link to get a new URL

!wget --no-check-certificate \
  "https://github.com/ADlead/Dogs-Cats/archive/master.zip" \
  -O "/tmp/cats-and-dogs.zip"

local_zip = '/tmp/cats-and-dogs.zip'
zip_ref = zipfile.ZipFile(local_zip, 'r')
zip_ref.extractall('/tmp')
zip_ref.close()

import cv2 as cv
import os
import numpy as np

import random
import pickle

import time

start_time = time.time()

data_dir = '/tmp/Dogs-Cats-master/data'
batch_save_path = '/tmp/Dogs-Cats-master/batch_files'

# 創(chuàng)建batch文件存儲的文件夾
os.makedirs(batch_save_path, exist_ok=True)

# 圖片統(tǒng)一大?。?00 * 100
# 訓(xùn)練集 20000：100個(gè)batch文件，每個(gè)文件200張圖片
# 驗(yàn)證集 5000： 一個(gè)測試文件，測試時(shí) 50張 x 100 批次

# 進(jìn)入圖片數(shù)據(jù)的目錄，讀取圖片信息
all_data_files = os.listdir(os.path.join(data_dir, 'train/'))

# print(all_data_files)

# 打算數(shù)據(jù)的順序
random.shuffle(all_data_files)

all_train_files = all_data_files[:20000]
all_test_files = all_data_files[20000:]

train_data = []
train_label = []
train_filenames = []

test_data = []
test_label = []
test_filenames = []

# 訓(xùn)練集
for each in all_train_files:
  img = cv.imread(os.path.join(data_dir,'train/',each),1)
  resized_img = cv.resize(img, (100,100))

  img_data = np.array(resized_img)
  train_data.append(img_data)
  if 'cat' in each:
    train_label.append(0)
  elif 'dog' in each:
    train_label.append(1)
  else:
    raise Exception('%s is wrong train file'%(each))
  train_filenames.append(each)

# 測試集
for each in all_test_files:
  img = cv.imread(os.path.join(data_dir,'train/',each), 1)
  resized_img = cv.resize(img, (100,100))

  img_data = np.array(resized_img)
  test_data.append(img_data)
  if 'cat' in each:
    test_label.append(0)
  elif 'dog' in each:
    test_label.append(1)
  else:
    raise Exception('%s is wrong test file'%(each))
  test_filenames.append(each)

print(len(train_data), len(test_data))

# 制作100個(gè)batch文件
start = 0
end = 200
for num in range(1, 101):
  batch_data = train_data[start: end]
  batch_label = train_label[start: end]
  batch_filenames = train_filenames[start: end]
  batch_name = 'training batch {} of 15'.format(num)

  all_data = {
    'data':batch_data,
    'label':batch_label,
    'filenames':batch_filenames,
    'name':batch_name
  }

  with open(os.path.join(batch_save_path, 'train_batch_{}'.format(num)), 'wb') as f:
    pickle.dump(all_data, f)

  start += 200
  end += 200

# 制作測試文件
all_test_data = {
  'data':test_data,
  'label':test_label,
  'filenames':test_filenames,
  'name':'test batch 1 of 1'
}

with open(os.path.join(batch_save_path, 'test_batch'), 'wb') as f:
  pickle.dump(all_test_data, f)


end_time = time.time()
print('制作結(jié)束, 用時(shí){}秒'.format(end_time - start_time))

import tensorflow as tf
import numpy as np
import cv2 as cv
import os
import pickle


''' 全局參數(shù) '''
IMAGE_SIZE = 100
LEARNING_RATE = 1e-4
TRAIN_STEP = 10000
TRAIN_SIZE = 100
TEST_STEP = 100
TEST_SIZE = 50

IS_TRAIN = True

SAVE_PATH = '/tmp/Dogs-Cats-master/model/'

data_dir = '/tmp/Dogs-Cats-master/batch_files'
pic_path = '/tmp/Dogs-Cats-master/data/test1'

''''''


def load_data(filename):
  '''從batch文件中讀取圖片信息'''
  with open(filename, 'rb') as f:
    data = pickle.load(f, encoding='iso-8859-1')
    return data['data'],data['label'],data['filenames']

# 讀取數(shù)據(jù)的類
class InputData:
  def __init__(self, filenames, need_shuffle):
    all_data = []
    all_labels = []
    all_names = []
    for file in filenames:
      data, labels, filename = load_data(file)

      all_data.append(data)
      all_labels.append(labels)
      all_names += filename

    self._data = np.vstack(all_data)
    self._labels = np.hstack(all_labels)
    print(self._data.shape)
    print(self._labels.shape)

    self._filenames = all_names

    self._num_examples = self._data.shape[0]
    self._need_shuffle = need_shuffle
    self._indicator = 0
    if self._indicator:
      self._shuffle_data()

  def _shuffle_data(self):
    # 把數(shù)據(jù)再混排
    p = np.random.permutation(self._num_examples)
    self._data = self._data[p]
    self._labels = self._labels[p]

  def next_batch(self, batch_size):
    '''返回每一批次的數(shù)據(jù)'''
    end_indicator = self._indicator + batch_size
    if end_indicator > self._num_examples:
      if self._need_shuffle:
        self._shuffle_data()
        self._indicator = 0
        end_indicator = batch_size
      else:
        raise Exception('have no more examples')
    if end_indicator > self._num_examples:
      raise Exception('batch size is larger than all examples')
    batch_data = self._data[self._indicator : end_indicator]
    batch_labels = self._labels[self._indicator : end_indicator]
    batch_filenames = self._filenames[self._indicator : end_indicator]
    self._indicator = end_indicator
    return batch_data, batch_labels, batch_filenames

# 定義一個(gè)類
class MyTensor:
  def __init__(self):


    # 載入訓(xùn)練集和測試集
    train_filenames = [os.path.join(data_dir, 'train_batch_%d'%i) for i in range(1, 101)]
    test_filenames = [os.path.join(data_dir, 'test_batch')]
    self.batch_train_data = InputData(train_filenames, True)
    self.batch_test_data = InputData(test_filenames, True)

    pass

  def flow(self):
    self.x = tf.placeholder(tf.float32, [None, IMAGE_SIZE, IMAGE_SIZE, 3], 'input_data')
    self.y = tf.placeholder(tf.int64, [None], 'output_data')
    self.keep_prob = tf.placeholder(tf.float32)

    # self.x = self.x / 255.0 需不需要這一步？

    # 圖片輸入網(wǎng)絡(luò)中
    fc = self.conv_net(self.x, self.keep_prob)

    self.loss = tf.losses.sparse_softmax_cross_entropy(labels=self.y, logits=fc)
    self.y_ = tf.nn.softmax(fc) # 計(jì)算每一類的概率
    self.predict = tf.argmax(fc, 1)
    self.acc = tf.reduce_mean(tf.cast(tf.equal(self.predict, self.y), tf.float32))

    self.train_op = tf.train.AdamOptimizer(LEARNING_RATE).minimize(self.loss)
    self.saver = tf.train.Saver(max_to_keep=1)

    print('計(jì)算流圖已經(jīng)搭建.')

  # 訓(xùn)練
  def myTrain(self):
    acc_list = []
    with tf.Session() as sess:
      sess.run(tf.global_variables_initializer())

      for i in range(TRAIN_STEP):
        train_data, train_label, _ = self.batch_train_data.next_batch(TRAIN_SIZE)

        eval_ops = [self.loss, self.acc, self.train_op]
        eval_ops_results = sess.run(eval_ops, feed_dict={
          self.x:train_data,
          self.y:train_label,
          self.keep_prob:0.7
        })
        loss_val, train_acc = eval_ops_results[0:2]

        acc_list.append(train_acc)
        if (i+1) % 100 == 0:
          acc_mean = np.mean(acc_list)
          print('step:{0},loss:{1:.5},acc:{2:.5},acc_mean:{3:.5}'.format(
            i+1,loss_val,train_acc,acc_mean
          ))
        if (i+1) % 1000 == 0:
          test_acc_list = []
          for j in range(TEST_STEP):
            test_data, test_label, _ = self.batch_test_data.next_batch(TRAIN_SIZE)
            acc_val = sess.run([self.acc],feed_dict={
              self.x:test_data,
              self.y:test_label,
              self.keep_prob:1.0
            })
            test_acc_list.append(acc_val)
          print('[Test ] step:{0}, mean_acc:{1:.5}'.format(
            i+1, np.mean(test_acc_list)
          ))
      # 保存訓(xùn)練后的模型
      os.makedirs(SAVE_PATH, exist_ok=True)
      self.saver.save(sess, SAVE_PATH + 'my_model.ckpt')

  def myTest(self):
    with tf.Session() as sess:
      model_file = tf.train.latest_checkpoint(SAVE_PATH)
      model = self.saver.restore(sess, save_path=model_file)
      test_acc_list = []
      predict_list = []
      for j in range(TEST_STEP):
        test_data, test_label, test_name = self.batch_test_data.next_batch(TEST_SIZE)
        for each_data, each_label, each_name in zip(test_data, test_label, test_name):
          acc_val, y__, pre, test_img_data = sess.run(
            [self.acc, self.y_, self.predict, self.x],
            feed_dict={
              self.x:each_data.reshape(1, IMAGE_SIZE, IMAGE_SIZE, 3),
              self.y:each_label.reshape(1),
              self.keep_prob:1.0
            }
          )
          predict_list.append(pre[0])
          test_acc_list.append(acc_val)

          # 把測試結(jié)果顯示出來
          self.compare_test(test_img_data, each_label, pre[0], y__[0], each_name)
      print('[Test ] mean_acc:{0:.5}'.format(np.mean(test_acc_list)))

  def compare_test(self, input_image_arr, input_label, output, probability, img_name):
    classes = ['cat', 'dog']
    if input_label == output:
      result = '正確'
    else:
      result = '錯(cuò)誤'
    print('測試【{0}】,輸入的label:{1}, 預(yù)測得是{2}:{3}的概率:{4:.5}, 輸入的圖片名稱:{5}'.format(
      result,input_label, output,classes[output], probability[output], img_name
    ))

  def conv_net(self, x, keep_prob):
    conv1_1 = tf.layers.conv2d(x, 16, (3, 3), padding='same', activation=tf.nn.relu, name='conv1_1')
    conv1_2 = tf.layers.conv2d(conv1_1, 16, (3, 3), padding='same', activation=tf.nn.relu, name='conv1_2')
    pool1 = tf.layers.max_pooling2d(conv1_2, (2, 2), (2, 2), name='pool1')

    conv2_1 = tf.layers.conv2d(pool1, 32, (3, 3), padding='same', activation=tf.nn.relu, name='conv2_1')
    conv2_2 = tf.layers.conv2d(conv2_1, 32, (3, 3), padding='same', activation=tf.nn.relu, name='conv2_2')
    pool2 = tf.layers.max_pooling2d(conv2_2, (2, 2), (2, 2), name='pool2')

    conv3_1 = tf.layers.conv2d(pool2, 64, (3, 3), padding='same', activation=tf.nn.relu, name='conv3_1')
    conv3_2 = tf.layers.conv2d(conv3_1, 64, (3, 3), padding='same', activation=tf.nn.relu, name='conv3_2')
    pool3 = tf.layers.max_pooling2d(conv3_2, (2, 2), (2, 2), name='pool3')

    conv4_1 = tf.layers.conv2d(pool3, 128, (3, 3), padding='same', activation=tf.nn.relu, name='conv4_1')
    conv4_2 = tf.layers.conv2d(conv4_1, 128, (3, 3), padding='same', activation=tf.nn.relu, name='conv4_2')
    pool4 = tf.layers.max_pooling2d(conv4_2, (2, 2), (2, 2), name='pool4')

    flatten = tf.layers.flatten(pool4) # 把網(wǎng)絡(luò)展平，以輸入到后面的全連接層

    fc1 = tf.layers.dense(flatten, 512, tf.nn.relu)
    fc1_dropout = tf.nn.dropout(fc1, keep_prob=keep_prob)
    fc2 = tf.layers.dense(fc1, 256, tf.nn.relu)
    fc2_dropout = tf.nn.dropout(fc2, keep_prob=keep_prob)
    fc3 = tf.layers.dense(fc2, 2, None) # 得到輸出fc3

    return fc3

  def main(self):
    self.flow()
    if IS_TRAIN is True:
      self.myTrain()
    else:
      self.myTest()

  def final_classify(self):
    all_test_files_dir = './data/test1'
    all_test_filenames = os.listdir(all_test_files_dir)
    if IS_TRAIN is False:
      self.flow()
      # self.classify()
      with tf.Session() as sess:
        model_file = tf.train.latest_checkpoint(SAVE_PATH)
        mpdel = self.saver.restore(sess,save_path=model_file)

        predict_list = []
        for each_filename in all_test_filenames:
          each_data = self.get_img_data(os.path.join(all_test_files_dir,each_filename))
          y__, pre, test_img_data = sess.run(
            [self.y_, self.predict, self.x],
            feed_dict={
              self.x:each_data.reshape(1, IMAGE_SIZE, IMAGE_SIZE, 3),
              self.keep_prob: 1.0
            }
          )
          predict_list.append(pre[0])
          self.classify(test_img_data, pre[0], y__[0], each_filename)

    else:
      print('now is training model...')

  def classify(self, input_image_arr, output, probability, img_name):
    classes = ['cat','dog']
    single_image = input_image_arr[0] #* 255
    if output == 0:
      output_dir = 'cat/'
    else:
      output_dir = 'dog/'
    os.makedirs(os.path.join('./classiedResult', output_dir), exist_ok=True)
    cv.imwrite(os.path.join('./classiedResult',output_dir, img_name),single_image)
    print('輸入的圖片名稱:{0}，預(yù)測得有{1:5}的概率是{2}:{3}'.format(
      img_name,
      probability[output],
      output,
      classes[output]
    ))

  # 根據(jù)名稱獲取圖片像素
  def get_img_data(self,img_name):
    img = cv.imread(img_name)
    resized_img = cv.resize(img, (100, 100))
    img_data = np.array(resized_img)

    return img_data




if __name__ == '__main__':

  mytensor = MyTensor()
  mytensor.main() # 用于訓(xùn)練或測試

  # mytensor.final_classify() # 用于最后的分類

  print('hello world')