{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "abcdefghijklmnopqrstuvwxyz0123456789\n"
     ]
    }
   ],
   "source": [
    "from captcha.image import ImageCaptcha\n",
    "from PIL import Image, ImageFont, ImageDraw\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import random\n",
    "import glob\n",
    "\n",
    "%matplotlib inline\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "\n",
    "import string\n",
    "characters = string.ascii_lowercase + string.digits # 验证码字符集合数字+英文\n",
    "# characters = string.digits + string.ascii_uppercase + string.ascii_lowercase # 验证码字符集合数字+英文\n",
    "# characters = string.digits # 验证码字符集合\n",
    "print(characters)\n",
    "\n",
    "fonts = ['/usr/share/fonts/WindowsFonts/fonts/ANTQUAB.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/AGENCYR.TTF',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/ANTQUABI.TTF',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/ARIALNI.TTF',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/Candara.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/cambriab.ttf',\n",
    "        '/usr/share/fonts/WindowsFonts/fonts/Candarai.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/calibri.ttf',\n",
    "        '/usr/share/fonts/WindowsFonts/fonts/constan.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/constanz.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/kaiu.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/simhei.ttf',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/STHUPO.TTF', # 粗体不要\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/STKAITI.TTF',\n",
    "        '/usr/share/fonts/WindowsFonts/fonts/STZHONGS.TTF']\n",
    "\n",
    "width, height, n_len, n_class = 200, 70, 6, len(characters) + 1  #图片宽、高，验证码最大长度，分类类别：字符集+1个空值"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    " # 防止 tensorflow 占用所有显存\n",
    "import tensorflow as tf\n",
    "import tensorflow.keras.backend as K\n",
    "\n",
    "config = tf.ConfigProto()\n",
    "config.gpu_options.allow_growth=True  #True \n",
    "sess = tf.Session(config=config)\n",
    "K.set_session(sess)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 定义 CTC Loss\n",
    "import tensorflow.keras.backend as K\n",
    "\n",
    "def ctc_lambda_func(args):\n",
    "    '''\n",
    "        定义ctc损失函数\n",
    "        参数：y_pred:预测值,labels:标签，input_length:lstm tiemstep,label_length:标签长度\n",
    "    '''    \n",
    "    y_pred, labels, input_length, label_length = args\n",
    "    return K.ctc_batch_cost(labels, y_pred, input_length, label_length)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 定义网络\n",
    "from tensorflow.keras.models import *\n",
    "from tensorflow.keras.layers import *\n",
    "\n",
    "input_tensor = Input((height, width, 3))\n",
    "x = input_tensor\n",
    "\n",
    "for i, n_cnn in enumerate([2, 2, 2, 2, 2]):   \n",
    "    for j in range(n_cnn):\n",
    "        x = Conv2D(32*2**min(i, 3), kernel_size=3, padding='same', kernel_initializer='he_uniform')(x)  # 32*2**min(i, 3)\n",
    "        x = BatchNormalization()(x)\n",
    "#         x = Activation('relu')(x) # 20200729 relu 改LeakyReLU\n",
    "        x = LeakyReLU(0.01)(x)\n",
    "    x = MaxPooling2D(2 if i < 4 else (2, 1))(x)\n",
    "#     tf.summary.image('conv2d', conv,10)\n",
    "    \n",
    "# def cnn_layer(index,inputs, filters, kernel_size, strides):\n",
    "#     x = Conv2D(filters, kernel_size=kernel_size, strides=strides[0], padding='same',name='cnn{}'.format(index + 1))(inputs)\n",
    "#     x = BatchNormalization(name='bn{}'.format(index + 1))(x)\n",
    "#     x = LeakyReLU(0.01)(x)\n",
    "#     x = MaxPooling2D(pool_size=(2,2), strides=strides[1], padding='same',name='pool{}'.format(index + 1))(x)\n",
    "#     return x\n",
    "# x = cnn_layer(0,inputs=x, kernel_size=7, filters=32, strides=(1, 1))\n",
    "# x = cnn_layer(1,inputs=x, kernel_size=5, filters=64, strides=(1, 2))\n",
    "# x = cnn_layer(2,inputs=x, kernel_size=3, filters=128, strides=(1, 2))\n",
    "# x = cnn_layer(3,inputs=x, kernel_size=3, filters=128, strides=(1, 2))\n",
    "# x = cnn_layer(4,inputs=x, kernel_size=3, filters=64, strides=(1, 2))\n",
    "\n",
    "x = Permute((2, 1, 3))(x)\n",
    "x = TimeDistributed(Flatten())(x)\n",
    "rnn_size = 128 # 128  32\n",
    "# x = Bidirectional(CuDNNGRU(rnn_size, return_sequences=True))(x)\n",
    "# x = Bidirectional(CuDNNGRU(rnn_size, return_sequences=True))(x)\n",
    "x = Bidirectional(GRU(rnn_size, return_sequences=True))(x)\n",
    "x = Bidirectional(GRU(rnn_size, return_sequences=True))(x)  # 200epoch 0.0153 - val_loss: 0.0136\n",
    "# x = Dropout(0.5)(x)\n",
    "x = Dense(n_class, activation='softmax')(x)\n",
    "\n",
    "base_model = Model(inputs=input_tensor, outputs=x)\n",
    "# base_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# labels = Input(name='the_labels', shape=[n_len], dtype='float32')\n",
    "labels = Input(name='the_labels', shape=[None], dtype='float32')\n",
    "input_length = Input(name='input_length', shape=[1], dtype='int64')\n",
    "label_length = Input(name='label_length', shape=[1], dtype='int64')\n",
    "loss_out = Lambda(ctc_lambda_func, output_shape=(1,), name='ctc')([x, labels, input_length, label_length])\n",
    "\n",
    "model = Model(inputs=[input_tensor, labels, input_length, label_length], outputs=loss_out)\n",
    "\n",
    "x= base_model.output    # [batch_sizes, series_length, classes]\n",
    "input_length = Input(batch_shape=[None], dtype='int32')\n",
    "ctc_decode = K.ctc_decode(x, input_length=input_length * K.shape(x)[1])\n",
    "decode = K.function([base_model.input, input_length], [ctc_decode[0][0]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # 网络结构可视化\n",
    "# from tensorflow.keras.utils import plot_model\n",
    "# from IPython.display import Image\n",
    "\n",
    "# plot_model(model, to_file='ctc.png', show_shapes=True)\n",
    "# Image('ctc.png')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 3600x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 279,
       "width": 2263
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# base_model.summary()\n",
    "def get_wavy_line(w = (0, 100),h = (30, 50)):\n",
    "    '''产生波浪线坐标'''\n",
    "    import random\n",
    "    n = 50\n",
    "    x = 0\n",
    "    y = random.randint(h[0],h[1])\n",
    "    flag = random.randint(0,2)\n",
    "    xy = [(x, y)]\n",
    "    while x < w[1]:\n",
    "        temp_y = random.randint(1, 3)\n",
    "        temp_x = random.randint(5, 10)\n",
    "        if flag == 0:\n",
    "            if y + temp_y > h[1]:\n",
    "                y -= temp_y\n",
    "                flag = 1\n",
    "            else:\n",
    "                y += temp_y\n",
    "        else:\n",
    "            if y - temp_y < h[0]:\n",
    "                y += temp_y\n",
    "                flag = 0\n",
    "            else:\n",
    "                y -= temp_y\n",
    "        x = x+temp_x if x+temp_x < w[1] else w[1]\n",
    "        xy.append((x, y))\n",
    "    return xy\n",
    "\n",
    "def gen_captcha(text, fig_size=(200,70), fonts=['/usr/share/fonts/WindowsFonts/fonts/ANTQUAB.TTF'],font_color=[(0,0,0)],same_color=1, font_size=(25, 35),offset_hor=5,offset_ver=5, rotate=0,\n",
    "                font_noise=0, line=(0,5), line_width=(1,2), point=(0,500),wavy=(0,0), noise_color=[(200,200,255)], bg=[(255,255,255)]):\n",
    "    '''\n",
    "    text:验证码文本\n",
    "    size:验证码图片宽高\n",
    "    fonts:字体列表，随机选择一个\n",
    "    font_noise: 字体散点干扰，0不加干扰，1加干扰\n",
    "    fill:字体颜色范围\n",
    "    rotate:字体旋转角度\n",
    "    line:干扰线条数范围\n",
    "    point:干扰点数范围\n",
    "    wavy:波浪线数范围\n",
    "    color:干扰线、点 颜色\n",
    "    bg:背景色范围\n",
    "    '''\n",
    "    bg = random.choice(bg)\n",
    "    img = Image.new(mode='RGB', size=fig_size, color=bg) #\n",
    "    draw = ImageDraw.Draw(im=img, mode='RGB') # im, mode=None\n",
    "    font = ImageFont.truetype(random.choice(fonts), size=random.randint(font_size[0], font_size[1])) # font=None, size=10, index=0, encoding=\"\"\n",
    "    rotate = random.randint(0, rotate)\n",
    "    def get_char_img(char,font,font_color,rotate,bg, font_noise=0):\n",
    "        '''\n",
    "        生成单个字符图片，随机颜色加随机旋转\n",
    "        \n",
    "        '''\n",
    "        w, h = draw.textsize(char, font=font)\n",
    "        im = Image.new('RGB',(w,h), color=bg)\n",
    "        ImageDraw.Draw(im).text((0,0), char, font=font, fill=font_color)   \n",
    "        if rotate:\n",
    "            im = im.rotate(random.randint(-rotate, rotate),Image.BILINEAR,expand=1)\n",
    "            im = im.crop(im.getbbox())\n",
    "        if font_noise: \n",
    "            im_draw = ImageDraw.Draw(im)\n",
    "#             for i in range(random.randint(10,100)):\n",
    "            for i in range(random.randint(int(w*h*0.01),int(w*h*0.05))):\n",
    "                im_draw.point(xy=(random.randint(0, w), random.randint(0, h)),fill=bg)\n",
    "        table = []\n",
    "        for i in range(256):\n",
    "            table.append(i * 97) # 5.97\n",
    "        mask = im.convert('L').point(table)         \n",
    "        return (im, mask)\n",
    "    \n",
    "    char_color = random.choice(font_color)\n",
    "    if same_color:        \n",
    "        char_imgs = [get_char_img(char, font, font_color=char_color, rotate=rotate, bg=bg, font_noise=font_noise) for char in text]\n",
    "    else:\n",
    "        char_imgs = [get_char_img(char, font, font_color=random.choice(font_color), rotate=rotate, bg=bg, font_noise=font_noise) for char in text]       \n",
    "    ws = [img[0].size[0] for img in char_imgs]\n",
    "    hs = [img[0].size[1] for img in char_imgs]\n",
    "    w = max(sum(ws), fig_size[0])\n",
    "    h = max(max(hs), fig_size[1])\n",
    "    if w>fig_size[0] or h>fig_size[1]:\n",
    "        img = img.resize((w+6, h+6), Image.BILINEAR)\n",
    "        draw = ImageDraw.Draw(im=img, mode='RGB')  # im, mode=None\n",
    "        fig_size = img.size\n",
    "\n",
    "    if rotate:\n",
    "        temp_x = random.randint(int((fig_size[0]-sum(ws))/5), int((fig_size[0]-sum(ws))/2+1))\n",
    "        temp_y = random.randint(int((fig_size[1]-hs[0])/8), int((fig_size[1]-hs[0])/2+1))\n",
    "        for i in range(len(char_imgs)):\n",
    "            img.paste(char_imgs[i][0], box=(temp_x, temp_y), mask=char_imgs[i][1]) \n",
    "            new_x = temp_x+ws[i]+random.randint(0, offset_hor)\n",
    "            temp_x = new_x if new_x<fig_size[0] else temp_x+ws[i]+random.randint(-offset_hor,0)\n",
    "            new_y = temp_y+random.randint(-offset_ver, offset_ver)\n",
    "            temp_y = new_y if 0<new_y and new_y+hs[i]<fig_size[1] else temp_y  \n",
    "        \n",
    "    # 直线\n",
    "    for i in range(random.randint(line[0], line[1])):\n",
    "        x0 = random.randint(0, fig_size[0])\n",
    "        x1 = random.randint(0, fig_size[0])\n",
    "        y0 = random.randint(0, fig_size[1])\n",
    "        y1 = random.randint(0, fig_size[1])\n",
    "        draw.line(xy=((x0,y0),(x1,y1)),\n",
    "                  fill=random.choice(noise_color),\n",
    "                  width=random.choice(line_width)) # xy, fill=None, width=0\n",
    "    # 散点\n",
    "    for i in range(random.randint(point[0], point[1])):\n",
    "        draw.point(xy=(random.randint(0, fig_size[0]), random.randint(0, fig_size[1])),\n",
    "                   fill=random.choice(noise_color))\n",
    "    # 波浪线\n",
    "    for _ in range(random.randint(wavy[0],wavy[1])):  \n",
    "        draw.line(xy=get_wavy_line(w = (0, 200),h = (min(hs)-5, max(hs)+5)), \n",
    "                  fill=char_color, width=random.choice(line_width))\n",
    "        \n",
    "    if not rotate:\n",
    "        temp_x = random.randint(int((fig_size[0]-sum(ws))/5), int((fig_size[0]-sum(ws))/2+1))\n",
    "        temp_y = random.randint(int((fig_size[1]-hs[0])/8), int((fig_size[1]-hs[0])/2+1))\n",
    "        for i in range(len(char_imgs)):\n",
    "            draw.text((temp_x, temp_y), text[i], font=font, fill=char_color)\n",
    "            new_x = temp_x+ws[i]+random.randint(0, offset_hor)\n",
    "            temp_x = new_x if new_x<fig_size[0] else temp_x+ws[i]+random.randint(-offset_hor,0)\n",
    "            new_y = temp_y+random.randint(-offset_ver, offset_ver)\n",
    "            temp_y = new_y if 0<new_y and new_y+hs[i]<fig_size[1] else temp_y             \n",
    "                \n",
    "    return img.resize((200,70), Image.BILINEAR)\n",
    "                   \n",
    "\n",
    "# img2 = Image.open('FileInfo0508/31c1f481-912a-11ea-b24d-408d5cd36814_cmftq.jpg') # 波浪线验证码\n",
    "# font_color=[(random.randint(210,255),random.randint(0,40),0),\n",
    "#             (0,random.randint(200,255),random.randint(0,90)),\n",
    "#            (random.randint(30,45),random.randint(90,110),255),\n",
    "#            (255,random.randint(210,255),random.randint(0,10)),\n",
    "#            (0,random.randint(0,5),random.randint(0,5))]\n",
    "# img = gen_captcha('cmftq', fig_size=(200,70), fonts=fonts,font_color=font_color,\n",
    "#                 same_color=1, font_size=(35, 45),offset_hor=8,offset_ver=5, rotate=5,\n",
    "#                 font_noise=1, line=(0,0), line_width=(1,2), point=(0,0),wavy=(1,1),\n",
    "#                 noise_color=[(200,200,255)], bg=[(255,255,255)])\n",
    "\n",
    "# img2 = Image.open('/data/captcha/shensebeijingsandian/pgv4_d58a8328-c425-11ea-be07-ecf4bbc56acd.jpg') # 深色背景验证码\n",
    "# font_color = [(random.randint(200,255),random.randint(0,40),random.randint(30,100)),\n",
    "#            (random.randint(150,200),random.randint(150,200),random.randint(150,200)),\n",
    "#            (random.randint(150,200),random.randint(150,250),random.randint(50,100))]\n",
    "# noise_color = [(random.randint(50,100),random.randint(150,250),random.randint(50,100)),\n",
    "#               (random.randint(200,250),random.randint(200,250),random.randint(50,100))]\n",
    "\n",
    "# img = gen_captcha('mtbd', fig_size=(100,25), fonts=fonts,font_color=font_color,same_color=1,\n",
    "#                   font_size=(15, 20),offset_hor=8,offset_ver=5, rotate=10,\n",
    "#                 font_noise=0, line=(0,0), line_width=(1,2), point=(30,100),wavy=(0,0), \n",
    "#                   noise_color=noise_color, bg=[(random.randint(20,100),random.randint(10,100),255)])\n",
    "\n",
    "\n",
    "\n",
    "# img2 = Image.open('/data/captcha/shensexiansandian/r4y6_f7bcd30f3c913228ba2404e83aea0806.jpg') #深色线斜体字\n",
    "# font_color = [(random.randint(10,50),random.randint(10,50),random.randint(100,160)),\n",
    "#              (random.randint(30,100),random.randint(30,50),random.randint(120,250))]\n",
    "# noise_color = [(random.randint(180,255),random.randint(10,50),random.randint(20,40)),\n",
    "#               (random.randint(10,20),random.randint(150,250),random.randint(0,40))]\n",
    "# img = gen_captcha('R1Y6', fig_size=(135,40), fonts=fonts,font_color=font_color,same_color=0, font_size=(20, 28),offset_hor=8,offset_ver=5, rotate=20,\n",
    "#                 font_noise=0, line=(2,8), line_width=(1,2,1,3,4), point=(20,200),wavy=(0,0), \n",
    "#                   noise_color=noise_color, bg=[(random.randint(220,255),random.randint(220,255),random.randint(220,255))])\n",
    "\n",
    "# img2 = Image.open('/data/captcha/0ad9.jpg').resize((200,70), Image.BILINEAR) #小图噪点 \n",
    "# font_color=[(random.randint(210,255),random.randint(0,40),0),\n",
    "#             (0,random.randint(200,255),random.randint(0,90)),\n",
    "#            (random.randint(30,45),random.randint(90,110),255),\n",
    "#            (255,random.randint(210,255),random.randint(0,10)),\n",
    "#            (0,random.randint(0,5),random.randint(0,5))]\n",
    "# noise_color=[(random.randint(210,255),random.randint(0,40),0),\n",
    "#             (0,random.randint(200,255),random.randint(0,90)),\n",
    "#            (random.randint(30,45),random.randint(90,110),255),\n",
    "#            (255,random.randint(210,255),random.randint(0,10)),\n",
    "#            (0,random.randint(0,5),random.randint(0,5))]\n",
    "# img = gen_captcha('0ade9', fig_size=(52,21), fonts=fonts,font_color=font_color,same_color=1, font_size=(15, 18),offset_hor=1,offset_ver=0, rotate=0,\n",
    "#                 font_noise=1, line=(0,0), line_width=(1,2,1), point=(20,100),wavy=(0,0), \n",
    "#                   noise_color=noise_color, bg=[(random.randint(220,255),random.randint(220,255),random.randint(220,255))])\n",
    "\n",
    "img2 = Image.open('/data/captcha/rph2.jpg').resize((200,70), Image.BILINEAR) #小图多种颜色字、干扰线 \n",
    "font_color=[(random.randint(60,90),random.randint(60,90),random.randint(60,90)),\n",
    "            (random.randint(90,120),random.randint(90,120),random.randint(90,120)),\n",
    "           (random.randint(120,150),random.randint(120,150),random.randint(120,150))]\n",
    "noise_color=[(random.randint(60,80),random.randint(60,90),random.randint(60,80)),\n",
    "            (random.randint(90,110),random.randint(90,110),random.randint(90,120)),\n",
    "           (random.randint(120,140),random.randint(120,140),random.randint(120,140))]\n",
    "img = gen_captcha('RPH2', fig_size=(70,26), fonts=fonts,font_color=font_color,same_color=0, font_size=(20, 24),offset_hor=1,offset_ver=1, rotate=0,\n",
    "                font_noise=1, line=(4,8), line_width=(1,1), point=(0,0),wavy=(0,0), \n",
    "                  noise_color=noise_color, bg=[(random.randint(220,255),random.randint(220,255),random.randint(220,255))])\n",
    "\n",
    "im = [img, img2]\n",
    "plt.figure(figsize=(50,10))\n",
    "for i in range(1,3):    \n",
    "    \n",
    "    plt.subplot(2,2,i)\n",
    "    plt.imshow(im[i-1])\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "label:  kaxkt\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fa5604249b0>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 150,
       "width": 370
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def random_xy(width,height):   \n",
    "    '''\n",
    "    随机位置函数，返回指定范围随机位置坐标\n",
    "    参数：width:图片宽，height:图片高\n",
    "    '''\n",
    "    x = random.randint(0, width)\n",
    "    y = random.randint(0, height)\n",
    "    return x, y\n",
    "def random_color(start, end, opacity=None):\n",
    "    '''\n",
    "    随机颜色函数，返回指定范围随机颜色值\n",
    "    参数：start:颜色最低值，end:颜色最高值\n",
    "    '''\n",
    "    red = random.randint(start, end)\n",
    "    green = random.randint(start, end)\n",
    "    blue = random.randint(start, end)\n",
    "    if opacity is None:\n",
    "        return (red, green, blue)\n",
    "    return (red, green, blue, opacity)\n",
    "# 重组验证码\n",
    "def rebuild_img(path):\n",
    "    '''\n",
    "    读取本地验证码图片进行随机加点噪声为新图片\n",
    "    参数：path：图片路径\n",
    "    返回：重组后图片    \n",
    "    '''\n",
    "    if re.search('FileInfo0508', path)!=None:\n",
    "        label = path.split('_')[-1][:-4].lower().replace('1','l')\n",
    "#         print(re.search('FileInfo0508', path))\n",
    "    elif re.search('/data/esa_sdk/gan/english', path)!=None:\n",
    "        label = path.split('_')[-1][:-4]\n",
    "    else:\n",
    "        label = path.split('/')[-1].split('_')[0]\n",
    "#         print('label',label)\n",
    "    crop_n = len(label) \n",
    "    img = Image.open(path)\n",
    "    img = img.convert('RGB')\n",
    "    width, height = img.size\n",
    "#     img2 = img2.resize((100,50), Image.BILINEAR)\n",
    "    draw = ImageDraw.Draw(img)\n",
    "    for _ in range(random.randint(0,450)):\n",
    "        draw.point(xy=(random_xy(width,height)),fill=random_color(25, 255))  \n",
    "    for _ in range(random.randint(0, 3)):\n",
    "        draw.line(xy=(random_xy(width, height),random_xy(width, height)), fill=random_color(20, 250), width=random.randint(1,2))\n",
    "    return img.resize((200,70), Image.NEAREST), label.lower()\n",
    "\n",
    "import re\n",
    "len4_imgs = []\n",
    "len5_imgs = []\n",
    "paths = 'FileInfo0508_2/*.jpg'\n",
    "paths = '/data/esa_sdk/gan/english/*.jpg'\n",
    "for path in glob.glob('/data/esa_sdk/gan/english/*.jpg')[:100]:\n",
    "    label = path.split('_')[-1][:-4].lower().replace('1','l')\n",
    "    if len(label) ==4:\n",
    "        len4_imgs.append(path)\n",
    "for path in glob.glob('/data/captcha/shensebeijingsandian/*.jpg')[:2500]:\n",
    "    label = path.split('_')[0].split('/')[-1].lower()\n",
    "    if len(label) ==4:\n",
    "        len4_imgs.append(path)\n",
    "for path in glob.glob('/data/captcha/shensexiansandian/*.jpg')[:2200]:\n",
    "    label = path.split('_')[0].split('/')[-1].lower()\n",
    "    if len(label) ==4:\n",
    "        len4_imgs.append(path)    \n",
    "\n",
    "for path in glob.glob('FileInfo0508_2/*.jpg')[:3000]:\n",
    "    label = path.split('_')[-1][:-4].lower().replace('1','l')\n",
    "    if len(label) ==5 and re.search('[0-9]', label)==None:\n",
    "        len5_imgs.append(path)   \n",
    "# for path in glob.glob('/data/captcha/kongxinbolang/*.jpg')[:1000]:\n",
    "#     label = path.split('_')[0].split('/')[-1].lower()\n",
    "#     if len(label) ==5:\n",
    "#         len5_imgs.append(path)   \n",
    "# random.shuffle(real_imgs)\n",
    "# img, l = rebuild_img(glob.glob('/data/esa_sdk/gan/english/*.jpg')[0])\n",
    "# img, l = rebuild_img(glob.glob('/data/captcha/shensexiansandian/*.jpg')[3])\n",
    "# img, l = rebuild_img(glob.glob('/data/captcha/dianxianduoyanse/*.jpg')[2])\n",
    "# img, l = rebuild_img(glob.glob('/data/captcha/shensebeijingsandian/*.jpg')[2])\n",
    "img, l = rebuild_img(glob.glob('FileInfo0508_2/*.jpg')[2])\n",
    "print('label: ',l)\n",
    "plt.imshow(img)\n",
    "# l2 = [rebuild_img(real_imgs[i])[1] for i in range(200,300)]\n",
    "# [characters.find(x) for x in l2[6]]\n",
    "# len5_imgs[-3:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/data/captcha/shensebeijingsandian/pgv4_d58a8328-c425-11ea-be07-ecf4bbc56acd.jpg pgv4\n"
     ]
    }
   ],
   "source": [
    "import re\n",
    "# real_imgs = []\n",
    "# paths = '/data/esa_sdk/gan/english/*.jpg' # 样本数 111 浅色英文数字 805bc78f-fbbd-11e9-9bc7-408d5cd36814_9tcl.jpg\n",
    "# paths = '/data/captcha/shensexiansandian/*.jpg' # 样本数：10006 深色线、斜体字、散点 r4y6_f7bcd30f3c913228ba2404e83aea0806.jpg\n",
    "# paths = 'FileInfo0508_2/*.jpg' # 样本数：3575 波浪线验证码  5789e596-9144-11ea-b24d-408d5cd36814_hqupb.jpg\n",
    "# paths = '/data/captcha/kongxinbolang/*.jpg'  # 样本数：1099 空心字验证码  u42sc_9a8399860afcbe11f50b51600630891c.jpg.\n",
    "paths = '/data/captcha/shensebeijingsandian/*.jpg' #样本数：2716 深色背景 pgv4_d58a8328-c425-11ea-be07-ecf4bbc56acd.jpg\n",
    "\n",
    "for path in glob.glob('/data/captcha/shensebeijingsandian/*.jpg')[:2500]:\n",
    "    label = path.split('_')[0].split('/')[-1].lower()\n",
    "    if len(label) ==4:\n",
    "        print(path, label)\n",
    "        break\n",
    "        \n",
    "# imgs = glob.glob(paths)\n",
    "# print(len(imgs), imgs[0])\n",
    "# plt.imshow(Image.open(imgs[0]))\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 定义数据生成器\n",
    "from tensorflow.keras.utils import Sequence\n",
    "\n",
    "class CaptchaSequence(Sequence):\n",
    "    '''\n",
    "    继承Sequence的数据生成类，方便调用多CPU，加快生成训练及测试数据\n",
    "    参数：self.characters:验证码字符集合，self.batch_size：每批次样本数，self.steps：生成多少批数据，self.n_len：验证码长度，\n",
    "    self.width：图片宽度，self.height:图片高度，self.input_length：lstm time step长度，self.label_length：标签长度\n",
    "    返回：array类型训练或测试数据 \n",
    "    \n",
    "    '''\n",
    "    def __init__(self, characters, batch_size, steps, n_len=6, width=200, height=70, \n",
    "                 input_length=12, label_length=6, chars_len=(5, 5)): #  width=128, height=64, input_length=16, label_length=4\n",
    "        self.characters = characters\n",
    "        self.batch_size = batch_size\n",
    "        self.steps = steps\n",
    "        self.n_len = n_len\n",
    "        self.width = width\n",
    "        self.height = height\n",
    "        self.input_length = input_length\n",
    "        self.label_length = label_length\n",
    "        self.chars_len = chars_len\n",
    "#         self.label_length = self.n_len\n",
    "        self.n_class = len(characters)+1\n",
    "#         self.n_class = -2\n",
    "        self.generator = ImageCaptcha(width=width, height=height, font_sizes=(12,20,18,25))\n",
    "#         self.fonts_list = glob.glob('/usr/share/fonts/WindowsFonts/fonts/*.ttf')\n",
    "        self.fonts_list = ['/usr/share/fonts/WindowsFonts/fonts/ANTQUAB.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/AGENCYR.TTF', # o显示像方框 \n",
    "         '/usr/share/fonts/WindowsFonts/fonts/ANTQUABI.TTF',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/ARIALNI.TTF',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/Candara.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/cambriab.ttf',\n",
    "        '/usr/share/fonts/WindowsFonts/fonts/Candarai.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/calibri.ttf',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/constan.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/constanz.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/kaiu.ttf',\n",
    "         '/usr/share/fonts/WindowsFonts/fonts/simhei.ttf',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/STHUPO.TTF',\n",
    "        '/usr/share/fonts/WindowsFonts/fonts/STKAITI.TTF',\n",
    "        '/usr/share/fonts/WindowsFonts/fonts/STZHONGS.TTF']\n",
    "#         self.fonts_list = ['/usr/share/fonts/WindowsFonts/fonts/arial.ttf','/usr/share/fonts/WindowsFonts/fonts/ANTQUAB.TTF',\n",
    "#                             '/usr/share/fonts/WindowsFonts/fonts/BKANT.TTF','/usr/share/fonts/WindowsFonts/fonts/cambriab.ttf',\n",
    "#                             '/usr/share/fonts/WindowsFonts/fonts/comic.ttf','/usr/share/fonts/WindowsFonts/fonts/GOTHIC.TTF']\n",
    "            \n",
    "    def __len__(self):\n",
    "        return self.steps\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        batch_label_length = random.choice([4,5])\n",
    "        self.n_len = n_len\n",
    "#         print('batch_label_length',batch_label_length)\n",
    "        X = np.zeros((self.batch_size, self.height, self.width, 3), dtype=np.float32)\n",
    "        y = np.zeros((self.batch_size, self.n_len), dtype=np.uint8)\n",
    "#         print(y)\n",
    "#         y = np.zeros((self.batch_size, batch_label_length), dtype=np.uint8)\n",
    "        input_length = np.ones(self.batch_size)*self.input_length\n",
    "        label_length = np.ones(self.batch_size)*self.n_len \n",
    "\n",
    "        for i in range(self.batch_size):\n",
    "#                 print('len 4',y.shape, i)\n",
    "            # 定义验证码字符集 (大写字母、小写字母、大写字母+数字)\n",
    "            gen_characters = random.choice([string.ascii_lowercase,string.ascii_lowercase+string.digits, string.ascii_uppercase, string.digits + string.ascii_uppercase])             \n",
    "            if i % 10 == 0:                     \n",
    "                image, random_str = rebuild_img(random.choice(len4_imgs)) \n",
    "            elif i % 10 == 1:\n",
    "                image, random_str = rebuild_img(random.choice(len5_imgs))  \n",
    "            elif i % 10 <= 3:                \n",
    "                random_str = ''.join([random.choice(gen_characters) for j in range(batch_label_length)])\n",
    "                font_color=[(random.randint(210,255),random.randint(0,40),0),\n",
    "                            (0,random.randint(200,255),random.randint(0,90)),\n",
    "                           (random.randint(30,45),random.randint(90,110),255),\n",
    "#                            (255,random.randint(210,255),random.randint(0,10)),\n",
    "                           (0,random.randint(0,5),random.randint(0,5))]\n",
    "                image = gen_captcha(random_str, fig_size=(200,70), fonts=fonts,font_color=font_color,\n",
    "                                same_color=1, font_size=(35, 45),offset_hor=8,offset_ver=5, rotate=5,\n",
    "                                font_noise=1, line=(0,0), line_width=(1,2), point=(0,0),wavy=(1,1),\n",
    "                                noise_color=[(200,200,255)], bg=[(255,255,255)])        \n",
    "#                 image = gen_captcha(random_str, size=(200,70), fonts=font,fill=(0,201),font_size=(30, 45), font_noise=1, rotate=(0,0),\n",
    "#                 line=(0,0), point=(0,0),wavy=(1,1), color=(0,255), bg=255)  # 产生波浪线干扰验证码               \n",
    "            elif i % 10 <= 5:\n",
    "                random_str = ''.join([random.choice(gen_characters) for j in range(batch_label_length)])\n",
    "                font_color = [(random.randint(200,255),random.randint(0,40),random.randint(30,100)),\n",
    "                           (random.randint(150,200),random.randint(150,200),random.randint(150,200)),\n",
    "                           (random.randint(150,200),random.randint(150,250),random.randint(50,100))]\n",
    "                noise_color = [(random.randint(50,100),random.randint(150,250),random.randint(50,100)),\n",
    "                              (random.randint(200,250),random.randint(200,250),random.randint(50,100))]\n",
    "\n",
    "                image = gen_captcha(random_str, fig_size=(100,25), fonts=fonts,font_color=font_color,same_color=1,\n",
    "                                  font_size=(15, 20),offset_hor=8,offset_ver=5, rotate=10,\n",
    "                                font_noise=0, line=(0,0), line_width=(1,2), point=(30,100),wavy=(0,0), \n",
    "                                  noise_color=noise_color, bg=[(random.randint(20,100),random.randint(10,100),255)])                \n",
    "#                 image = gen_captcha(random_str, size=(100,25), fonts=fonts,fill=(150,200),font_size=(15, 20), font_noise=0, rotate=(0,0),\n",
    "#                 line=(0,0), point=(50,150),wavy=(0,0), color=(150,200), bg=57)  # 深色背景\n",
    "            elif i%10<=7:\n",
    "                random_str = ''.join([random.choice(gen_characters) for j in range(batch_label_length)])\n",
    "                font_color = [(random.randint(50,100),random.randint(10,50),random.randint(100,250)),\n",
    "                             (random.randint(0,50),random.randint(0,50),random.randint(120,250))]\n",
    "                noise_color = [(random.randint(200,225),random.randint(0,50),random.randint(20,100)),\n",
    "                              (random.randint(50,120),random.randint(150,250),random.randint(50,80))]\n",
    "#                 font_color = [(random.randint(10,50),random.randint(10,50),random.randint(100,160)),\n",
    "#                              (random.randint(30,100),random.randint(30,50),random.randint(120,250))]\n",
    "#                 noise_color = [(random.randint(180,255),random.randint(10,50),random.randint(20,40)),\n",
    "#                               (random.randint(10,20),random.randint(150,250),random.randint(0,40))]\n",
    "                image = gen_captcha(random_str, fig_size=(135,40), fonts=fonts,font_color=font_color,same_color=0, font_size=(20, 28),offset_hor=8,offset_ver=5, rotate=10,\n",
    "                                font_noise=0, line=(2,8), line_width=(1,2,1,3,4), point=(20,200),wavy=(0,0), \n",
    "                                  noise_color=noise_color, bg=[(random.randint(220,255),random.randint(220,255),random.randint(220,255))])            \n",
    "#                 image = gen_captcha(random_str, size=(135,40), fonts=fonts,fill=(20,60),font_size=(20, 28), font_noise=0, rotate=(-6,6),\n",
    "#                 line=(2,5),line_width=(1,3), point=(50,150),wavy=(0,0), color=(80,170), bg=255)   #深色线斜体字  \n",
    "            elif i%10<=8:\n",
    "                random_str = ''.join([random.choice(gen_characters) for j in range(batch_label_length)])\n",
    "                font_color=[(random.randint(60,90),random.randint(60,90),random.randint(60,90)),\n",
    "                            (random.randint(90,120),random.randint(90,120),random.randint(90,120)),\n",
    "                           (random.randint(120,150),random.randint(120,150),random.randint(120,150))]\n",
    "                noise_color=[(random.randint(60,80),random.randint(60,90),random.randint(60,80)),\n",
    "                            (random.randint(90,110),random.randint(90,110),random.randint(90,120)),\n",
    "                           (random.randint(120,140),random.randint(120,140),random.randint(120,140))]\n",
    "                image = gen_captcha(random_str, fig_size=(70,26), fonts=fonts,font_color=font_color,same_color=0, font_size=(20, 24),offset_hor=1,offset_ver=1, rotate=0,\n",
    "                                font_noise=1, line=(4,8), line_width=(1,1), point=(0,0),wavy=(0,0), \n",
    "                                  noise_color=noise_color, bg=[(random.randint(220,255),random.randint(220,255),random.randint(220,255))])\n",
    "            else:\n",
    "                random_str = ''.join([random.choice(gen_characters) for j in range(batch_label_length)])\n",
    "                font_color=[(random.randint(210,255),random.randint(0,40),0),\n",
    "                            (0,random.randint(200,255),random.randint(0,90)),\n",
    "                           (random.randint(30,45),random.randint(90,110),255),                           \n",
    "                           (0,random.randint(0,5),random.randint(0,5))]\n",
    "                noise_color=[(random.randint(210,255),random.randint(0,40),0),\n",
    "                            (0,random.randint(200,255),random.randint(0,90)),\n",
    "                           (random.randint(30,45),random.randint(90,110),255),\n",
    "                           (255,random.randint(210,255),random.randint(0,10)),\n",
    "                           (0,random.randint(0,5),random.randint(0,5))]\n",
    "                image = gen_captcha(random_str, fig_size=(52,21), fonts=fonts,font_color=font_color,same_color=1, font_size=(15, 18),offset_hor=1,offset_ver=0, rotate=0,\n",
    "                                font_noise=1, line=(0,0), line_width=(1,2,1), point=(20,100),wavy=(0,0), \n",
    "                                  noise_color=noise_color, bg=[(random.randint(220,255),random.randint(220,255),random.randint(220,255))])\n",
    "#             X[i] = np.expand_dims(np.array(image)/255.0, axis=-1)\n",
    "            X[i] = np.array(image)/255.0\n",
    "            label = [self.characters.find(x) for x in random_str.lower()] # 全部标签转为小写\n",
    "            if len(random_str) < self.n_len:\n",
    "                label += [self.n_class]*(self.n_len-len(random_str))                \n",
    "            y[i] = label\n",
    "        return [X, y, input_length, label_length], np.ones(self.batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'tlmnt')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 163,
       "width": 370
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# # 测试生成器\n",
    "# data = CaptchaSequence(characters, batch_size=20,n_len=5, width=200, height=70, steps=1, chars_len=(4, 5))\n",
    "# [X_test, y_test, input_length, label_length], _  = data[0]\n",
    "idx =10\n",
    "# plt.imshow(X_test[idx][:,:,0])\n",
    "plt.imshow(X_test[idx])\n",
    "plt.title(''.join([characters[x] for x in y_test[idx] if x < len(characters)]))\n",
    "# print(input_length, label_length)\n",
    "# # print(y_test)\n",
    "# # print(X_test.shape)\n",
    "# print(n_class)\n",
    "# print(y_test[idx])\n",
    "# print(y_test)\n",
    "# characters\n",
    "# print(id(X_test))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 30, 80, 3)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fec11600358>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 158,
       "width": 370
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 从现有图片生成测试数据\n",
    "def get_data(img_path):\n",
    "    img = Image.open(img_path)\n",
    "#     img = img.crop((0, height-25, width, height))\n",
    "    w, h = img.size\n",
    "    data = np.zeros((1,h, w, 3))\n",
    "    data[0] = np.array(img)/255.0\n",
    "    return data\n",
    "img_path = '../FileInfo/ffc510f4-f977-11e9-b970-408d5cd36814_5802.jpg'\n",
    "\n",
    "data = get_data(img_path)\n",
    "print(data.shape)\n",
    "plt.imshow(data[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 准确率回调函数\n",
    "from tqdm import tqdm\n",
    "\n",
    "def evaluate(model, batch_size=128, steps=1):\n",
    "    '''\n",
    "    准确率验证函数，每批次的验证码长度必须一致\n",
    "    '''    \n",
    "    batch_acc = 0\n",
    "    valid_data = CaptchaSequence(characters, batch_size, steps)\n",
    "    for i in range(len(valid_data)):\n",
    "        [X_test, y_test, _, _], _ = valid_data[i]\n",
    "        y_pred = base_model.predict(X_test)\n",
    "        shape = y_pred.shape\n",
    "    #         out = K.get_value(K.ctc_decode(y_pred, input_length=np.ones(shape[0])*shape[1],)[0][0])[:, :4]\n",
    "        out = K.get_value(K.ctc_decode(y_pred, input_length=np.ones(shape[0])*shape[1],)[0][0])[:, :]\n",
    "    #     print(y_test)\n",
    "    #     print(type(y_test))\n",
    "    #     print(y_test[y_test<10, axis=1])\n",
    "    #     print(out)\n",
    "        if out.shape[1] >= 4:\n",
    "            batch_acc += (y_test[:,:out.shape[1]] == out).all(axis=1).mean()\n",
    "    return batch_acc / steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.27000000000000002"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# model.load_weights('digit4to6_ctc_best2.h5')\n",
    "evaluate(base_model,batch_size=1, steps=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import Callback\n",
    "\n",
    "class Evaluate(Callback):\n",
    "    '''\n",
    "    准确率验证的类，每批次的验证码长度必须一致\n",
    "    '''    \n",
    "    def __init__(self):\n",
    "        self.accs = []\n",
    "    \n",
    "    def on_epoch_end(self, epoch, logs=None):\n",
    "        logs = logs or {}\n",
    "        acc = evaluate(base_model, batch_size=128)  # evaluate(base_model)\n",
    "        logs['val_acc'] = acc\n",
    "        self.accs.append(acc)\n",
    "        print('\\nacc%.4f'%acc)\n",
    "#         print(f'\\nacc: {acc*100:.4f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input_1 True\n",
      "conv2d True\n",
      "batch_normalization True\n",
      "activation True\n",
      "conv2d_1 True\n",
      "batch_normalization_1 True\n",
      "activation_1 True\n",
      "max_pooling2d True\n"
     ]
    }
   ],
   "source": [
    "model.load_weights('gru_english4to6_ctc_best.h5')\n",
    "# 前8层不训练\n",
    "for layer in model.layers[:8]:\n",
    "    layer.trainable = True\n",
    "    print(layer.name, layer.trainable)\n",
    "# del train_data\n",
    "# del valid_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "51670\n",
      "Epoch 1/100\n",
      " 999/1000 [============================>.] - ETA: 0s - loss: 0.3002Epoch 1/100\n",
      "1000/1000 [==============================] - 366s 366ms/step - loss: 0.3001 - val_loss: 0.3828\n",
      "Epoch 2/100\n",
      "1000/1000 [==============================] - 335s 335ms/step - loss: 0.2433 - val_loss: 0.3172\n",
      "Epoch 3/100\n",
      "1000/1000 [==============================] - 336s 336ms/step - loss: 0.1989 - val_loss: 0.2927\n",
      "Epoch 4/100\n",
      "1000/1000 [==============================] - 333s 333ms/step - loss: 0.1782 - val_loss: 0.2895\n",
      "Epoch 5/100\n",
      "1000/1000 [==============================] - 346s 346ms/step - loss: 0.1565 - val_loss: 0.2390\n",
      "Epoch 6/100\n",
      "1000/1000 [==============================] - 320s 320ms/step - loss: 0.1445 - val_loss: 0.2038\n",
      "Epoch 7/100\n",
      "1000/1000 [==============================] - 307s 307ms/step - loss: 0.1286 - val_loss: 0.2020\n",
      "Epoch 8/100\n",
      "1000/1000 [==============================] - 350s 350ms/step - loss: 0.1182 - val_loss: 0.1584\n",
      "Epoch 9/100\n",
      "1000/1000 [==============================] - 328s 328ms/step - loss: 0.1167 - val_loss: 0.3541\n",
      "Epoch 10/100\n",
      "1000/1000 [==============================] - 315s 315ms/step - loss: 0.1115 - val_loss: 0.1415\n",
      "Epoch 11/100\n",
      "1000/1000 [==============================] - 305s 305ms/step - loss: 0.1015 - val_loss: 0.1296\n",
      "Epoch 12/100\n",
      "1000/1000 [==============================] - 348s 348ms/step - loss: 0.0975 - val_loss: 0.1122\n",
      "Epoch 13/100\n",
      "1000/1000 [==============================] - 324s 324ms/step - loss: 0.0986 - val_loss: 0.2596\n",
      "Epoch 14/100\n",
      "1000/1000 [==============================] - 309s 309ms/step - loss: 0.0912 - val_loss: 0.1162\n",
      "Epoch 15/100\n",
      "1000/1000 [==============================] - 299s 299ms/step - loss: 0.0914 - val_loss: 0.1713\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x7fddc982c630>"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluate()\n",
    "# 模型训练\n",
    "from tensorflow.keras.callbacks import EarlyStopping, CSVLogger, ModelCheckpoint\n",
    "from tensorflow.keras.optimizers import *\n",
    "import gc \n",
    "# del train_data\n",
    "# del valid_data\n",
    "# print(gc.collect())\n",
    "model.load_weights('gru_english4to6_ctc_best_5.h5')\n",
    "\n",
    "train_data = CaptchaSequence(characters, batch_size=128, steps=1000,input_length=12, label_length=6,chars_len=(5, 5)) # (characters, batch_size=128, steps=1000)\n",
    "valid_data = CaptchaSequence(characters, batch_size=128, steps=100,input_length=12, label_length=6,chars_len=(5, 5))  # (characters, batch_size=128, steps=100)\n",
    "# callbacks = [EarlyStopping(patience=5), Evaluate(), \n",
    "#              CSVLogger('ctc.csv'), ModelCheckpoint('ctc_best.h5', save_best_only=True)]\n",
    "callbacks = [EarlyStopping(patience=3),ModelCheckpoint('gru_english4to6_ctc_best_5.h5', save_best_only=True)]\n",
    "model.compile(loss={'ctc': lambda y_true, y_pred: y_pred}, optimizer=Adam(1e-3, amsgrad=True))\n",
    "model.fit_generator(train_data, epochs=100, validation_data=valid_data, workers=4, use_multiprocessing=True,\n",
    "                   callbacks=callbacks)\n",
    "# del train_data\n",
    "# del valid_data\n",
    "# print(gc.collect())\n",
    "# train_data = CaptchaSequence(characters, batch_size=128, steps=1000,input_length=12, label_length=5) # (characters, batch_size=128, steps=1000)\n",
    "# valid_data = CaptchaSequence(characters, batch_size=128, steps=100,input_length=12, label_length=5)  # (characters, batch_size=128, steps=100)\n",
    "# # 载入最好的模型继续训练一会\n",
    "# model.load_weights('gru_english4to6_ctc_best.h5')\n",
    "# # callbacks = [EarlyStopping(patience=5),\n",
    "# #              CSVLogger('ctc.csv', append=True), ModelCheckpoint('ctc_best.h5', save_best_only=True)]\n",
    "# callbacks = [CSVLogger('ctc.csv', append=True), ModelCheckpoint('gru_english4to6_ctc_best.h5', save_best_only=True)]\n",
    "\n",
    "# model.compile(loss={'ctc': lambda y_true, y_pred: y_pred}, optimizer=Adam(1e-4, amsgrad=True))\n",
    "# model.fit_generator(train_data, epochs=200, validation_data=valid_data, workers=4, use_multiprocessing=True,\n",
    "#                     callbacks=callbacks)\n",
    "\n",
    "# gru_english4to6_ctc_best_3.h5  loss: 0.0308 - val_loss: 0.0214 效果不好\n",
    "# gru_english4to6_ctc_best_6.h5 20200727 rgb图单gru 运行12轮 从loss: 6.2998 - val_loss: 1.0465降到 loss: 0.1202 - val_loss: 0.1727\n",
    "# gru_english4to6_ctc_best_5.h5 leakyReLU 双gru 7轮 loss: 0.1672 - val_loss: 0.8717\n",
    "# gru_english4to6_ctc_best_5.h5 没加组合图前LeakyReLU 双gru 再运行15轮loss: 0.0914 - val_loss: 0.1713"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "174\n",
      "Epoch 1/300\n",
      " 999/1000 [============================>.] - ETA: 0s - loss: 0.0246\n",
      "1000/1000 [==============================] - 303s 303ms/step - loss: 0.0245 - val_loss: 0.0286\n",
      "Epoch 2/300\n",
      "1000/1000 [==============================] - 286s 286ms/step - loss: 0.0246 - val_loss: 0.0324\n",
      "Epoch 3/300\n",
      "1000/1000 [==============================] - 288s 288ms/step - loss: 0.0241 - val_loss: 0.0246\n",
      "Epoch 4/300\n",
      "1000/1000 [==============================] - 308s 308ms/step - loss: 0.0231 - val_loss: 0.0270\n",
      "Epoch 5/300\n",
      "1000/1000 [==============================] - 292s 292ms/step - loss: 0.0232 - val_loss: 0.0258\n",
      "Epoch 6/300\n",
      "1000/1000 [==============================] - 299s 299ms/step - loss: 0.0221 - val_loss: 0.0305\n",
      "Epoch 7/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0229 - val_loss: 0.0208\n",
      "Epoch 8/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0208 - val_loss: 0.0271\n",
      "Epoch 9/300\n",
      "1000/1000 [==============================] - 289s 289ms/step - loss: 0.0233 - val_loss: 0.0279\n",
      "Epoch 10/300\n",
      "1000/1000 [==============================] - 291s 291ms/step - loss: 0.0226 - val_loss: 0.0290\n",
      "Epoch 11/300\n",
      "1000/1000 [==============================] - 285s 285ms/step - loss: 0.0227 - val_loss: 0.0249\n",
      "Epoch 12/300\n",
      "1000/1000 [==============================] - 296s 296ms/step - loss: 0.0213 - val_loss: 0.0238\n",
      "Epoch 13/300\n",
      "1000/1000 [==============================] - 290s 290ms/step - loss: 0.0208 - val_loss: 0.0270\n",
      "Epoch 14/300\n",
      "1000/1000 [==============================] - 292s 292ms/step - loss: 0.0217 - val_loss: 0.0204\n",
      "Epoch 15/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0203 - val_loss: 0.0262\n",
      "Epoch 16/300\n",
      "1000/1000 [==============================] - 285s 285ms/step - loss: 0.0221 - val_loss: 0.0249\n",
      "Epoch 17/300\n",
      "1000/1000 [==============================] - 288s 288ms/step - loss: 0.0212 - val_loss: 0.0260\n",
      "Epoch 18/300\n",
      "1000/1000 [==============================] - 294s 294ms/step - loss: 0.0220 - val_loss: 0.0279\n",
      "Epoch 19/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0203 - val_loss: 0.0314\n",
      "Epoch 20/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0211 - val_loss: 0.0286\n",
      "Epoch 21/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0210 - val_loss: 0.0301\n",
      "Epoch 22/300\n",
      "1000/1000 [==============================] - 287s 287ms/step - loss: 0.0205 - val_loss: 0.0308\n",
      "Epoch 23/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0201 - val_loss: 0.0244\n",
      "Epoch 24/300\n",
      "1000/1000 [==============================] - 291s 291ms/step - loss: 0.0195 - val_loss: 0.0276\n",
      "Epoch 25/300\n",
      "1000/1000 [==============================] - 293s 293ms/step - loss: 0.0200 - val_loss: 0.0246\n",
      "Epoch 26/300\n",
      "1000/1000 [==============================] - 290s 290ms/step - loss: 0.0209 - val_loss: 0.0283\n",
      "Epoch 27/300\n",
      "1000/1000 [==============================] - 289s 289ms/step - loss: 0.0210 - val_loss: 0.0290\n",
      "Epoch 28/300\n",
      "1000/1000 [==============================] - 288s 288ms/step - loss: 0.0203 - val_loss: 0.0222\n",
      "Epoch 29/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0200 - val_loss: 0.0222\n",
      "Epoch 30/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0212 - val_loss: 0.0316\n",
      "Epoch 31/300\n",
      "1000/1000 [==============================] - 295s 295ms/step - loss: 0.0200 - val_loss: 0.0211\n",
      "Epoch 32/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0206 - val_loss: 0.0350\n",
      "Epoch 33/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0206 - val_loss: 0.0270\n",
      "Epoch 34/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0199 - val_loss: 0.0395\n",
      "Epoch 35/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0202 - val_loss: 0.0481\n",
      "Epoch 36/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0204 - val_loss: 0.0308\n",
      "Epoch 37/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0210 - val_loss: 0.0253\n",
      "Epoch 38/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0198 - val_loss: 0.0245\n",
      "Epoch 39/300\n",
      "1000/1000 [==============================] - 292s 292ms/step - loss: 0.0206 - val_loss: 0.0248\n",
      "Epoch 40/300\n",
      "1000/1000 [==============================] - 296s 296ms/step - loss: 0.0204 - val_loss: 0.0283\n",
      "Epoch 41/300\n",
      "1000/1000 [==============================] - 288s 288ms/step - loss: 0.0212 - val_loss: 0.0255\n",
      "Epoch 42/300\n",
      "1000/1000 [==============================] - 287s 287ms/step - loss: 0.0197 - val_loss: 0.0247\n",
      "Epoch 43/300\n",
      "1000/1000 [==============================] - 289s 289ms/step - loss: 0.0206 - val_loss: 0.0311\n",
      "Epoch 44/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0200 - val_loss: 0.0327\n",
      "Epoch 45/300\n",
      "1000/1000 [==============================] - 289s 289ms/step - loss: 0.0194 - val_loss: 0.0330\n",
      "Epoch 46/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0197 - val_loss: 0.0243\n",
      "Epoch 47/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0196 - val_loss: 0.0222\n",
      "Epoch 48/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0203 - val_loss: 0.0263\n",
      "Epoch 49/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0194 - val_loss: 0.0280\n",
      "Epoch 50/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0197 - val_loss: 0.0262\n",
      "Epoch 51/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0196 - val_loss: 0.0290\n",
      "Epoch 52/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0186 - val_loss: 0.0309\n",
      "Epoch 53/300\n",
      "1000/1000 [==============================] - 287s 287ms/step - loss: 0.0188 - val_loss: 0.0223\n",
      "Epoch 54/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0191 - val_loss: 0.0244\n",
      "Epoch 55/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0186 - val_loss: 0.0257\n",
      "Epoch 56/300\n",
      "1000/1000 [==============================] - 321s 321ms/step - loss: 0.0194 - val_loss: 0.0225\n",
      "Epoch 57/300\n",
      "1000/1000 [==============================] - 301s 301ms/step - loss: 0.0198 - val_loss: 0.0290\n",
      "Epoch 58/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0189 - val_loss: 0.0263\n",
      "Epoch 59/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0189 - val_loss: 0.0231\n",
      "Epoch 60/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0193 - val_loss: 0.0244\n",
      "Epoch 61/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0186 - val_loss: 0.0276\n",
      "Epoch 62/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0187 - val_loss: 0.0200\n",
      "Epoch 63/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0174 - val_loss: 0.0217\n",
      "Epoch 64/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0191 - val_loss: 0.0223\n",
      "Epoch 65/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0208 - val_loss: 0.0259\n",
      "Epoch 66/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0188 - val_loss: 0.0256\n",
      "Epoch 67/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0184 - val_loss: 0.0181\n",
      "Epoch 68/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0177 - val_loss: 0.0272\n",
      "Epoch 69/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0180 - val_loss: 0.0327\n",
      "Epoch 70/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0183 - val_loss: 0.0319\n",
      "Epoch 71/300\n",
      "1000/1000 [==============================] - 279s 279ms/step - loss: 0.0175 - val_loss: 0.0213\n",
      "Epoch 72/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0172 - val_loss: 0.0297\n",
      "Epoch 73/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0187 - val_loss: 0.0306\n",
      "Epoch 74/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0175 - val_loss: 0.0268\n",
      "Epoch 75/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0171 - val_loss: 0.0252\n",
      "Epoch 76/300\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0196 - val_loss: 0.0276\n",
      "Epoch 77/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0176 - val_loss: 0.0290\n",
      "Epoch 78/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0188 - val_loss: 0.0209\n",
      "Epoch 79/300\n",
      "1000/1000 [==============================] - 284s 284ms/step - loss: 0.0189 - val_loss: 0.0272\n",
      "Epoch 80/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0185 - val_loss: 0.0263\n",
      "Epoch 81/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0187 - val_loss: 0.0271\n",
      "Epoch 82/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0186 - val_loss: 0.0240\n",
      "Epoch 83/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0176 - val_loss: 0.0223\n",
      "Epoch 84/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0181 - val_loss: 0.0220\n",
      "Epoch 85/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0167 - val_loss: 0.0196\n",
      "Epoch 86/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0181 - val_loss: 0.0195\n",
      "Epoch 87/300\n",
      "1000/1000 [==============================] - 279s 279ms/step - loss: 0.0182 - val_loss: 0.0272\n",
      "Epoch 88/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0183 - val_loss: 0.0229\n",
      "Epoch 89/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0176 - val_loss: 0.0260\n",
      "Epoch 90/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0173 - val_loss: 0.0303\n",
      "Epoch 91/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0179 - val_loss: 0.0272\n",
      "Epoch 92/300\n",
      "1000/1000 [==============================] - 279s 279ms/step - loss: 0.0184 - val_loss: 0.0212\n",
      "Epoch 93/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0183 - val_loss: 0.0233\n",
      "Epoch 94/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0174 - val_loss: 0.0277\n",
      "Epoch 95/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0170 - val_loss: 0.0229\n",
      "Epoch 96/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0173 - val_loss: 0.0246\n",
      "Epoch 97/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0173 - val_loss: 0.0209\n",
      "Epoch 98/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0177 - val_loss: 0.0302\n",
      "Epoch 99/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0182 - val_loss: 0.0298\n",
      "Epoch 100/300\n",
      "1000/1000 [==============================] - 279s 279ms/step - loss: 0.0183 - val_loss: 0.0212\n",
      "Epoch 101/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0181 - val_loss: 0.0261\n",
      "Epoch 102/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0178 - val_loss: 0.0287\n",
      "Epoch 103/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0183 - val_loss: 0.0292\n",
      "Epoch 104/300\n",
      "1000/1000 [==============================] - 285s 285ms/step - loss: 0.0180 - val_loss: 0.0192\n",
      "Epoch 105/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0165 - val_loss: 0.0285\n",
      "Epoch 106/300\n",
      "1000/1000 [==============================] - 294s 294ms/step - loss: 0.0176 - val_loss: 0.0228\n",
      "Epoch 107/300\n",
      "1000/1000 [==============================] - 291s 291ms/step - loss: 0.0168 - val_loss: 0.0204\n",
      "Epoch 108/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0177 - val_loss: 0.0229\n",
      "Epoch 109/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0190 - val_loss: 0.0240\n",
      "Epoch 110/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0179 - val_loss: 0.0210\n",
      "Epoch 111/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0186 - val_loss: 0.0257\n",
      "Epoch 112/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0178 - val_loss: 0.0194\n",
      "Epoch 113/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0175 - val_loss: 0.0196\n",
      "Epoch 114/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0173 - val_loss: 0.0214\n",
      "Epoch 115/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0180 - val_loss: 0.0199\n",
      "Epoch 116/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0170 - val_loss: 0.0302\n",
      "Epoch 117/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0165 - val_loss: 0.0294\n",
      "Epoch 118/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0166 - val_loss: 0.0210\n",
      "Epoch 119/300\n",
      "1000/1000 [==============================] - 279s 279ms/step - loss: 0.0180 - val_loss: 0.0188\n",
      "Epoch 120/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0167 - val_loss: 0.0189\n",
      "Epoch 121/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0175 - val_loss: 0.0222\n",
      "Epoch 122/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0164 - val_loss: 0.0213\n",
      "Epoch 123/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0161 - val_loss: 0.0240\n",
      "Epoch 124/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0162 - val_loss: 0.0201\n",
      "Epoch 125/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0169 - val_loss: 0.0219\n",
      "Epoch 126/300\n",
      "1000/1000 [==============================] - 279s 279ms/step - loss: 0.0175 - val_loss: 0.0193\n",
      "Epoch 127/300\n",
      "1000/1000 [==============================] - 285s 285ms/step - loss: 0.0169 - val_loss: 0.0553\n",
      "Epoch 128/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0168 - val_loss: 0.0257\n",
      "Epoch 129/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0170 - val_loss: 0.0245\n",
      "Epoch 130/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0175 - val_loss: 0.0265\n",
      "Epoch 131/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0165 - val_loss: 0.0195\n",
      "Epoch 132/300\n",
      "1000/1000 [==============================] - 281s 281ms/step - loss: 0.0163 - val_loss: 0.0187\n",
      "Epoch 133/300\n",
      "1000/1000 [==============================] - 280s 280ms/step - loss: 0.0180 - val_loss: 0.0241\n",
      "Epoch 134/300\n",
      "1000/1000 [==============================] - 279s 279ms/step - loss: 0.0170 - val_loss: 0.0211\n",
      "Epoch 135/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0157 - val_loss: 0.0180\n",
      "Epoch 136/300\n",
      "1000/1000 [==============================] - 282s 282ms/step - loss: 0.0171 - val_loss: 0.0242\n",
      "Epoch 137/300\n",
      "1000/1000 [==============================] - 285s 285ms/step - loss: 0.0165 - val_loss: 0.0189\n",
      "Epoch 138/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0171 - val_loss: 0.0210\n",
      "Epoch 139/300\n",
      "1000/1000 [==============================] - 286s 286ms/step - loss: 0.0170 - val_loss: 0.0182\n",
      "Epoch 140/300\n",
      "1000/1000 [==============================] - 283s 283ms/step - loss: 0.0166 - val_loss: 0.0270\n",
      "Epoch 141/300\n",
      "1000/1000 [==============================] - 289s 289ms/step - loss: 0.0172 - val_loss: 0.0197\n",
      "Epoch 142/300\n",
      "1000/1000 [==============================] - 291s 291ms/step - loss: 0.0163 - val_loss: 0.0221\n",
      "Epoch 143/300\n",
      "1000/1000 [==============================] - 289s 289ms/step - loss: 0.0159 - val_loss: 0.0277\n",
      "Epoch 144/300\n",
      "1000/1000 [==============================] - 290s 290ms/step - loss: 0.0157 - val_loss: 0.0207\n",
      "Epoch 145/300\n",
      "1000/1000 [==============================] - 291s 291ms/step - loss: 0.0170 - val_loss: 0.0244\n",
      "Epoch 146/300\n",
      "1000/1000 [==============================] - 289s 289ms/step - loss: 0.0159 - val_loss: 0.0186\n",
      "Epoch 147/300\n",
      "1000/1000 [==============================] - 292s 292ms/step - loss: 0.0166 - val_loss: 0.0278\n",
      "Epoch 148/300\n",
      "1000/1000 [==============================] - 304s 304ms/step - loss: 0.0159 - val_loss: 0.0303\n",
      "Epoch 149/300\n",
      "1000/1000 [==============================] - 310s 310ms/step - loss: 0.0152 - val_loss: 0.0199\n",
      "Epoch 150/300\n",
      "1000/1000 [==============================] - 309s 309ms/step - loss: 0.0172 - val_loss: 0.0208\n",
      "Epoch 151/300\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000/1000 [==============================] - 307s 307ms/step - loss: 0.0163 - val_loss: 0.0216\n",
      "Epoch 152/300\n",
      "1000/1000 [==============================] - 297s 297ms/step - loss: 0.0168 - val_loss: 0.0327\n",
      "Epoch 153/300\n",
      "1000/1000 [==============================] - 320s 320ms/step - loss: 0.0151 - val_loss: 0.0224\n",
      "Epoch 154/300\n",
      "1000/1000 [==============================] - 342s 342ms/step - loss: 0.0168 - val_loss: 0.0210\n",
      "Epoch 155/300\n",
      "  37/1000 [>.............................] - ETA: 4:22 - loss: 0.0184"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Process ForkPoolWorker-1837:\n",
      "Process ForkPoolWorker-1839:\n",
      "Process ForkPoolWorker-1838:\n",
      "Process ForkPoolWorker-1840:\n",
      "Traceback (most recent call last):\n",
      "Traceback (most recent call last):\n",
      "Traceback (most recent call last):\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/process.py\", line 254, in _bootstrap\n",
      "    self.run()\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/process.py\", line 254, in _bootstrap\n",
      "    self.run()\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/process.py\", line 254, in _bootstrap\n",
      "    self.run()\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/process.py\", line 93, in run\n",
      "    self._target(*self._args, **self._kwargs)\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/process.py\", line 93, in run\n",
      "    self._target(*self._args, **self._kwargs)\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/process.py\", line 254, in _bootstrap\n",
      "    self.run()\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/process.py\", line 93, in run\n",
      "    self._target(*self._args, **self._kwargs)\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/pool.py\", line 108, in worker\n",
      "    task = get()\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/pool.py\", line 119, in worker\n",
      "    result = (True, func(*args, **kwds))\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/process.py\", line 93, in run\n",
      "    self._target(*self._args, **self._kwargs)\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/queues.py\", line 342, in get\n",
      "    with self._rlock:\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/pool.py\", line 108, in worker\n",
      "    task = get()\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/pool.py\", line 108, in worker\n",
      "    task = get()\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/synchronize.py\", line 96, in __enter__\n",
      "    return self._semlock.__enter__()\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/queues.py\", line 343, in get\n",
      "    res = self._reader.recv_bytes()\n",
      "KeyboardInterrupt\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/connection.py\", line 216, in recv_bytes\n",
      "    buf = self._recv_bytes(maxlength)\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/site-packages/tensorflow/python/keras/utils/data_utils.py\", line 432, in get_index\n",
      "    return _SHARED_SEQUENCES[uid][i]\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/queues.py\", line 342, in get\n",
      "    with self._rlock:\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/connection.py\", line 407, in _recv_bytes\n",
      "    buf = self._recv(4)\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/synchronize.py\", line 96, in __enter__\n",
      "    return self._semlock.__enter__()\n",
      "  File \"<ipython-input-9-8e468a0f7e0f>\", line 108, in __getitem__\n",
      "    noise_color=noise_color, bg=[(random.randint(220,255),random.randint(220,255),random.randint(220,255))])\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/multiprocessing/connection.py\", line 379, in _recv\n",
      "    chunk = read(handle, remaining)\n",
      "KeyboardInterrupt\n",
      "  File \"<ipython-input-7-c349a12ee008>\", line 103, in gen_captcha\n",
      "    width=random.choice(line_width)) # xy, fill=None, width=0\n",
      "KeyboardInterrupt\n",
      "  File \"/home/python/anaconda3/envs/dl_nlp/lib/python3.5/site-packages/PIL/ImageDraw.py\", line 158, in line\n",
      "    if joint == \"curve\" and width > 4:\n",
      "KeyboardInterrupt\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-13-6ad8915c1054>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     16\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcompile\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mloss\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m{\u001b[0m\u001b[0;34m'ctc'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mlambda\u001b[0m \u001b[0my_true\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my_pred\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0my_pred\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mAdam\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1e-4\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mamsgrad\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     17\u001b[0m model.fit_generator(train_data, epochs=300, validation_data=valid_data, workers=4, use_multiprocessing=True,\n\u001b[0;32m---> 18\u001b[0;31m                     callbacks=callbacks)\n\u001b[0m\u001b[1;32m     19\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     20\u001b[0m \u001b[0;31m# 20200724 'gru_english4to6_ctc_best.h5' 灰度图 单lstm 运行236轮 loss: 0.0190 - val_loss: 0.0201\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/dl_nlp/lib/python3.5/site-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(self, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)\u001b[0m\n\u001b[1;32m   1777\u001b[0m         \u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0muse_multiprocessing\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1778\u001b[0m         \u001b[0mshuffle\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mshuffle\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1779\u001b[0;31m         initial_epoch=initial_epoch)\n\u001b[0m\u001b[1;32m   1780\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1781\u001b[0m   def evaluate_generator(self,\n",
      "\u001b[0;32m~/anaconda3/envs/dl_nlp/lib/python3.5/site-packages/tensorflow/python/keras/engine/training_generator.py\u001b[0m in \u001b[0;36mfit_generator\u001b[0;34m(model, generator, steps_per_epoch, epochs, verbose, callbacks, validation_data, validation_steps, class_weight, max_queue_size, workers, use_multiprocessing, shuffle, initial_epoch)\u001b[0m\n\u001b[1;32m    202\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    203\u001b[0m         outs = model.train_on_batch(\n\u001b[0;32m--> 204\u001b[0;31m             x, y, sample_weight=sample_weight, class_weight=class_weight)\n\u001b[0m\u001b[1;32m    205\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    206\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mouts\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/dl_nlp/lib/python3.5/site-packages/tensorflow/python/keras/engine/training.py\u001b[0m in \u001b[0;36mtrain_on_batch\u001b[0;34m(self, x, y, sample_weight, class_weight)\u001b[0m\n\u001b[1;32m   1550\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1551\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_train_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1552\u001b[0;31m       \u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain_function\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mins\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1553\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1554\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/dl_nlp/lib/python3.5/site-packages/tensorflow/python/keras/backend.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, inputs)\u001b[0m\n\u001b[1;32m   2912\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_callable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfeed_arrays\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfeed_symbols\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msymbol_vals\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msession\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2913\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2914\u001b[0;31m     \u001b[0mfetched\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_callable_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0marray_vals\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2915\u001b[0m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call_fetch_callbacks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfetched\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_fetches\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2916\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0mfetched\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/envs/dl_nlp/lib/python3.5/site-packages/tensorflow/python/client/session.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1380\u001b[0m           ret = tf_session.TF_SessionRunCallable(\n\u001b[1;32m   1381\u001b[0m               \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_session\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_handle\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstatus\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1382\u001b[0;31m               run_metadata_ptr)\n\u001b[0m\u001b[1;32m   1383\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mrun_metadata\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1384\u001b[0m           \u001b[0mproto_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtf_session\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTF_GetBuffer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrun_metadata_ptr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping, CSVLogger, ModelCheckpoint\n",
    "from tensorflow.keras.optimizers import *\n",
    "# del train_data\n",
    "# del valid_data\n",
    "# del data\n",
    "import gc \n",
    "print(gc.collect())\n",
    "# 载入最好的模型继续训练一会\n",
    "model.load_weights('gru_english4to6_ctc_best_5.h5')\n",
    "train_data = CaptchaSequence(characters, batch_size=128, steps=1000) # (characters, batch_size=128, steps=1000)\n",
    "valid_data = CaptchaSequence(characters, batch_size=128, steps=100)  # (characters, batch_size=128, steps=100)\n",
    "# callbacks = [EarlyStopping(patience=5),\n",
    "#              CSVLogger('ctc.csv', append=True), ModelCheckpoint('ctc_best.h5', save_best_only=True)]\n",
    "callbacks = [CSVLogger('ctc.csv', append=True), ModelCheckpoint('gru_english4to6_ctc_best_5.h5', save_best_only=True)]\n",
    "\n",
    "model.compile(loss={'ctc': lambda y_true, y_pred: y_pred}, optimizer=Adam(1e-4, amsgrad=True))\n",
    "model.fit_generator(train_data, epochs=300, validation_data=valid_data, workers=4, use_multiprocessing=True,\n",
    "                    callbacks=callbacks)\n",
    "\n",
    "# 20200724 'gru_english4to6_ctc_best.h5' 灰度图 单lstm 运行236轮 loss: 0.0190 - val_loss: 0.0201\n",
    "#gru_english4to6_ctc_best_6.h5 单gru运行300轮 loss: 0.0310 - val_loss: 0.0321 loss: 0.0261 - val_loss: 0.0282\n",
    "#gru_english4to6_ctc_best_6.h5 彩图加小图后再训练 loss: 0.0308 - val_loss: 0.0304 效果较好，取为最后结果\n",
    "# gru_english4to6_ctc_best_5.h5 LeakyReLU双gru 209轮  loss: 0.0201 - val_loss: 0.0224"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# model.load_weights('ctc_best.h5')\n",
    "# model.load_weights('gru_english4to6_ctc_best.h5')\n",
    "# base_model.save('gru_english_base_model5.h5')\n",
    "# base_model.load_weights('gru_english_base_model5.h5')   \n",
    "# model.load_weights('gru_english4to6_ctc_best_6.h5')\n",
    "base_model.save('gru_english_base_model0730.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "pred:aquoq\n",
      "true:aquqq\n",
      "1\n",
      "总耗时： 67.25510096549988\n",
      "正确数：3274， 错误数：1\n"
     ]
    }
   ],
   "source": [
    "# 测试模型\n",
    "# model.load_weights('gru_english4to6_ctc_best_3.h5') # 22层网络单lstm32 正确数：247， 错误数：1753 loss: 0.0116 - val_loss: 0.0071\n",
    "# model.load_weights('gru_english4to6_ctc_best5.h5')  # 22层网络单lstm128 正确数：1977， 错误数：23 loss: 0.0300 - val_loss: 0.0197\n",
    "# model.load_weights('gru_english4to6_ctc_best.h5') # 23层网络双lstm，正确数：324， 错误数：1676\n",
    "# model.load_weights('gru_english4to6_ctc_best_2.h5') # 19层网络 \n",
    "# model.load_weights('gru_english4to6_ctc_best4.h5') # 22层网络单lstm32 正确数：997， 错误数：1003\n",
    "# model.load_weights('gru_english4to6_ctc_best_5.h5') # 22层网络单lstm32\n",
    "characters2 = characters + ' '\n",
    "import time\n",
    "import re\n",
    "def get_test_data():\n",
    "    '''\n",
    "    从本地获取验证码图片并生成测试数据\n",
    "    '''    \n",
    "    X = []\n",
    "    Y = []\n",
    "#     for path in glob.glob('/data/captcha/shensexiansandian/*.jpg')[0:5000]:  # kongxinbolang/*.jpg  shensexiansandian/*.jpg\n",
    "#         random_str = path.split('_')[0].split('/')[-1].lower()  # ('/data/captcha/shensebeijingsandian/*.jpg')[:2500]:\n",
    "#         if 1:\n",
    "                                              \n",
    "    for path in glob.glob('FileInfo0508_2/*.jpg')[300:]: #FileInfo0508_2/*.jpg  '/data/esa_sdk/gan/english/*.jpg'\n",
    "        random_str = path.split('_')[-1][:-4].replace('1', 'l') \n",
    "        if len(random_str) ==5 and re.search('[0-9]', random_str)==None:\n",
    "            random_str = random_str.lower()\n",
    "\n",
    "#     for path in glob.glob('/data/esa_sdk/gan/english/*.jpg')[:]:     \n",
    "#         random_str = path.split('_')[-1][:-4]\n",
    "#         if len(random_str) ==4:                     \n",
    "#             random_str = random_str.lower()\n",
    "            \n",
    "#             print(random_str)\n",
    "#         if random_str.isdigit() and len(random_str) ==4:\n",
    "            img = Image.open(path)\n",
    "            img = img.convert('RGB')\n",
    "#             img = img.convert('L')\n",
    "            img = img.resize((200,70), Image.NEAREST)\n",
    "            X.append(np.array(img)/255.0)\n",
    "#             X.append(np.expand_dims(np.array(img)/255.0, axis=-1))\n",
    "            label_idx = [characters.find(x) for x in random_str]\n",
    "#             if len(random_str) < n_len:\n",
    "#                 label_idx += [n_class-1]*(n_len-len(random_str)) \n",
    "            Y.append(label_idx)\n",
    "    in_len = np.ones(len(X))*int(base_model.outputs[0].shape[1])\n",
    "    lab_len = np.ones(len(Y))*len(random_str)\n",
    "    return [np.array(X), np.array(Y), in_len, lab_len],np.ones(len(X))\n",
    "\n",
    "data = [get_test_data()]\n",
    "# \n",
    "# data = CaptchaSequence(characters, batch_size=200, steps=5, chars_len=(6,6))\n",
    "# \n",
    "\n",
    "\n",
    "pos = neg = 0\n",
    "t1 = time.time()\n",
    "err_img = []\n",
    "err_label = []\n",
    "for i in range(len(data)): \n",
    "    flag = False\n",
    "    [X_test, y_test, input_len, label_len], _  = data[i]\n",
    "    for idx in range(len(X_test)):\n",
    "        in_data = X_test[idx:idx+1]\n",
    "        out_pre = decode([in_data, np.ones(in_data.shape[0])])\n",
    "#         print(out_pre)\n",
    "        out = ''.join([characters[x] for x in out_pre[0][0]])  \n",
    "        \n",
    "#         y_pred = base_model.predict(X_test[idx:idx+1])\n",
    "#         out_pre = K.get_value(K.ctc_decode(y_pred, input_length=np.ones(y_pred.shape[0])*y_pred.shape[1])[0][0])[:, :6 ]\n",
    "#         out = ''.join([characters[x] for x in out_pre[0]]) \n",
    "        \n",
    "#         print(len( tf.get_default_graph().as_graph_def().node) )\n",
    "     \n",
    "        y_true = ''.join([characters[x] for x in y_test[idx] if x < len(characters)])\n",
    "#         print('out', out, y_true)\n",
    "        if out != y_true:\n",
    "            err_img.append(X_test[idx])\n",
    "            err_label.append('pre: %s, lab: %s'%(out, y_true))\n",
    "            print('pred:' + str(out) + '\\ntrue:' + str(y_true))\n",
    "            neg += 1\n",
    "            flag = True\n",
    "        else:\n",
    "            pos += 1 \n",
    "print(len(err_img))\n",
    "t2 = time.time()\n",
    "print('总耗时：',t2-t1)\n",
    "print('正确数：%d， 错误数：%d'%(pos,neg))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "12726\n",
      "36866\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x1296 with 15 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 617,
       "width": 1040
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(len( tf.get_default_graph().as_graph_def().node) )\n",
    "del data\n",
    "print(gc.collect())\n",
    "\n",
    "# paths = glob.glob('FileInfo0508_2/*.jpg')[:20] #FileInfo0508_2/*.jpg  '/data/esa_sdk/gan/english/*.jpg'\n",
    "# img = Image.open(paths[0])\n",
    "# img_l = img.convert('L')\n",
    "# print(img.mode)\n",
    "# print(img_l.mode)\n",
    "# imgs = [img, img_l]\n",
    "# fig = plt.figure(figsize=(18,18))\n",
    "# for i in range(2):\n",
    "#     plt.subplot(2,2,i+1)\n",
    "#     plt.imshow(imgs[i])\n",
    "# plt.show()\n",
    "# img = img.convert('RGB')\n",
    "\n",
    "# d0 = img.getdata()\n",
    "# d1 = img_l.getdata()\n",
    "# R,G,B = d0[5]\n",
    "# print(d1[5],int(R*0.299 + G*0.587 + B*0.114), (R*19595 + G*38469 + B*7472) >> 16)\n",
    "# Gray = R*0.299 + G*0.587 + B*0.114    \n",
    "# Gray = (R*19595 + G*38469 + B*7472) >> 16\n",
    "# err_img = []\n",
    "# for path in glob.glob('/data/captcha/shensebeijingsandian/*.jpg')[2500:2520]:  # kongxinbolang/*.jpg  shensexiansandian/*.jpg\n",
    "#     random_str = path.split('_')[0].split('/')[-1].lower()  # ('/data/captcha/shensebeijingsandian/*.jpg')[:2500]:\n",
    "#     if 1:\n",
    "#         img = Image.open(path)\n",
    "#         img = img.convert('L')\n",
    "#         img = img.resize((200,70), Image.NEAREST)\n",
    "#         print(path)\n",
    "#         err_img.append(img)\n",
    "#         err_img.append(np.expand_dims(np.array(img)/255.0, axis=-1))\n",
    "\n",
    "fig = plt.figure(figsize=(18,18))\n",
    "n = len(err_img[:20])\n",
    "for i in range(n):\n",
    "    plt.subplot(int(n/5)+1,5,i+1)\n",
    "    plt.title(err_label[i])\n",
    "#     plt.imshow(err_img[i][:,:,0])\n",
    "    plt.imshow(err_img[i])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x1080 with 9 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 721,
       "width": 872
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import requests\n",
    "import time\n",
    "from io import BytesIO\n",
    "# url=\"https://www.holdcg.com/servlet/validateCodeServlet?1595923859770\"\n",
    "url='http://zc.yatai.com/MetaPortlet/EPP/LoginPortlet/ConfigTools/randomcode.ashx?secode=ecisp_seccode&0.9867346532958197'\n",
    "\n",
    "test_img = []\n",
    "test_label = []\n",
    "for i in range(9):    \n",
    "    response=requests.get(url)\n",
    "    img = response.content\n",
    "    time.sleep(0.1)\n",
    "    image = Image.open(BytesIO(img))\n",
    "    image = image.convert('RGB')\n",
    "    img_arr = np.array(image.resize((200,70), Image.NEAREST))/255.0\n",
    "    x_test = np.array([img_arr])\n",
    "#     with graph.as_default():\n",
    "    pre = decode([x_test, np.ones(x_test.shape[0])])[0]\n",
    "    out = ''.join([characters[x] for x in pre[0]])  \n",
    "    test_img.append(image)\n",
    "    test_label.append(out)\n",
    "#         with open('/data/captcha/request_img/'+out+'_'+str(i)+'.jpg', 'wb') as f:\n",
    "#             f.write(img)\n",
    "plt.figure(figsize=(15,15))\n",
    "for i in range(len(test_img)):\n",
    "    plt.subplot(3,3,i+1)\n",
    "    plt.imshow(test_img[i])\n",
    "    plt.title(test_label[i])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 199,
   "metadata": {},
   "outputs": [],
   "source": [
    "evaluate(base_model,batch_size=128, steps=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f61db44d4e0>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 展示损失下降图\n",
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv('ctc.csv')\n",
    "df[['loss', 'val_loss']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "20\n"
     ]
    }
   ],
   "source": [
    "import re\n",
    "num = 0\n",
    "for path in glob.glob('FileInfo0508/*.jpg'): #    Digit5/*.jpg  ../FileInfo1031/*.jpg\n",
    "    random_str = path.split('_')[-1][:-4] \n",
    "    if re.search('[a-zA-Z]', random_str) != None and len(random_str) ==4:\n",
    "        num += 1\n",
    "print(num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# from PIL import Image, ImageFont, ImageDraw\n",
    "\n",
    "# def random_color(start, end, opacity=None):\n",
    "#     '''\n",
    "#     随机颜色函数，返回指定范围随机颜色值\n",
    "#     参数：start:颜色最低值，end:颜色最高值\n",
    "#     '''\n",
    "#     red = random.randint(start, end)\n",
    "#     green = random.randint(start, end)\n",
    "#     blue = random.randint(start, end)\n",
    "#     if opacity is None:\n",
    "#         return (red, green, blue)\n",
    "#     return (red, green, blue, opacity)\n",
    "# def random_xy(width,height):   \n",
    "#     '''\n",
    "#     随机位置函数，返回指定范围随机位置坐标\n",
    "#     参数：width:图片宽，height:图片高\n",
    "#     '''\n",
    "#     x = random.randint(0, width)\n",
    "#     y = random.randint(0, height)\n",
    "#     return x, y\n",
    "\n",
    "# table  =  []\n",
    "# for  i  in  range( 256 ):\n",
    "#     table.append( i * 1.97 )\n",
    "\n",
    "# def get_wavy_line(w = (0, 100),h = (30, 50)):\n",
    "#     '''产生波浪线坐标'''\n",
    "#     import random\n",
    "#     n = 50\n",
    "#     x = 0\n",
    "#     y = random.randint(h[0],h[1])\n",
    "#     flag = random.randint(0,2)\n",
    "#     xy = [(x, y)]\n",
    "#     while x < w[1]:\n",
    "#         temp_y = random.randint(1, 3)\n",
    "#         temp_x = random.randint(5, 10)\n",
    "#         if flag == 0:\n",
    "#             if y + temp_y > h[1]:\n",
    "#                 y -= temp_y\n",
    "#                 flag = 1\n",
    "#             else:\n",
    "#                 y += temp_y\n",
    "#         else:\n",
    "#             if y - temp_y < h[0]:\n",
    "#                 y += temp_y\n",
    "#                 flag = 0\n",
    "#             else:\n",
    "#                 y -= temp_y\n",
    "#         x = x+temp_x if x+temp_x < w[1] else w[1]\n",
    "#         xy.append((x, y))\n",
    "#     return xy\n",
    "\n",
    "# def get_wavy_text(n = 5,h = (25, 40)):\n",
    "#     import random\n",
    "#     y = random.randint(h[0],h[1])\n",
    "#     flag = random.randint(0,2)\n",
    "#     t_h = []\n",
    "#     for i in range(n):\n",
    "#         temp_y = random.randint(1, 5)\n",
    "#         temp_x = random.randint(20, 38)\n",
    "#         if flag == 0:\n",
    "#             if y + temp_y > h[1]:\n",
    "#                 y -= temp_y\n",
    "#                 flag = 1\n",
    "#             else:\n",
    "#                 y += temp_y\n",
    "#         else:\n",
    "#             if y - temp_y < h[0]:\n",
    "#                 y += temp_y\n",
    "#                 flag = 0\n",
    "#             else:\n",
    "#                 y -= temp_y\n",
    "#         t_h.append(y)\n",
    "#     return t_h\n",
    "\n",
    "# def generate_image(chars, background = random_color(255,255), width=200, height=70,char_color = [(213,0,0),(255,214,0),(41,98,255),(0,200,83),(0,0,0)], fonts =None):\n",
    "#     '''\n",
    "#     生成验证码图片\n",
    "#     chars：要生成的字符串\n",
    "#     background:背景颜色\n",
    "#     '''\n",
    "#     image = Image.new('RGB', (width, height), color=background)\n",
    "#     draw = ImageDraw.Draw(image)\n",
    "#     def get_char_img(char,font,color,angle):\n",
    "#         '''\n",
    "#         生成单个字符图片，随机颜色加随机旋转\n",
    "        \n",
    "#         '''\n",
    "#         w, h = draw.textsize(char, font=font)\n",
    "# #         w, h = ImageDraw.Draw.textsize(char, font=font)\n",
    "#         im = Image.new('RGBA',(w,h), color=background)\n",
    "#         ImageDraw.Draw(im).text((0,0), char, font=font, fill=color)\n",
    "#         im = im.crop(im.getbbox())\n",
    "#         rot = im.rotate(angle,Image.BILINEAR,expand=1)\n",
    "#         bg = Image.new('RGBA',rot.size,background)\n",
    "#         im = Image.composite(rot, bg, rot)\n",
    "#         return im\n",
    "#     w_all = random.randint(10,30)\n",
    "#     im_list = []\n",
    "#     w_list = []\n",
    "#     fonts = ['/usr/share/fonts/WindowsFonts/fonts/ANTQUAB.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/AGENCYR.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/ANTQUABI.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/ARIALNI.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/Candara.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/cambriab.ttf',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/Candarai.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/calibri.ttf',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/constan.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/constanz.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/kaiu.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/simhei.ttf',\n",
    "# #         '/usr/share/fonts/WindowsFonts/fonts/STHUPO.TTF',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/STKAITI.TTF',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/STZHONGS.TTF']\n",
    "# #     fonts = ['/usr/share/fonts/WindowsFonts/fonts/ANTQUAB.TTF','/usr/share/fonts/WindowsFonts/fonts/cambriab.ttf',\n",
    "# #             '/usr/share/fonts/WindowsFonts/fonts/Candarai.ttf','/usr/share/fonts/WindowsFonts/fonts/calibri.ttf']\n",
    "#     font = ImageFont.truetype(font=random.choice(fonts) , size=random.randint(33,35))\n",
    "\n",
    "# #     f = '/usr/share/fonts/WindowsFonts/fonts/calibri.ttf' # font=random.choice(fonts)\n",
    "# #     font = ImageFont.truetype(font=f , size=random.randint(33,35))\n",
    "    \n",
    "#     ch_color = random.choice(char_color+[random_color(0,255)])\n",
    "#     max_h = height\n",
    "#     for c in chars:\n",
    "# #         fonts = ['/usr/share/fonts/WindowsFonts/fonts/ariali.ttf','/usr/share/fonts/WindowsFonts/fonts/arial.ttf',\n",
    "# #                 '/usr/share/fonts/WindowsFonts/fonts/cambriab.ttf','/usr/share/fonts/WindowsFonts/fonts/verdana.ttf']\n",
    "# #         font = ImageFont.truetype(font=random.choice(fonts), size=random.randint(18,25))\n",
    "#         char_img = get_char_img(char=c, font=font, color=ch_color, angle=random.randint(-10,10))\n",
    "#         w, h = char_img.size\n",
    "#         max_h = h if h > max_h else max_h\n",
    "#         w_all += random.randint(-4,1)  \n",
    "#         w_list.append(w_all)\n",
    "#         im_list.append(char_img)\n",
    "# #         image.paste(char_img, (w_all,random.randint(0,image.size[1]-h)))\n",
    "#         w_all += w\n",
    "#     w_all = width if w_all < width else w_all\n",
    "# #     if w_all > width or max_h>height:\n",
    "#     image = Image.new('RGB', (w_all, max_h), color=background)\n",
    "#     draw = ImageDraw.Draw(image)\n",
    "    \n",
    "#     temp_h = 0\n",
    "#     t_h = get_wavy_text(n = 6,h = (10, 30))\n",
    "#     for i in range(len(w_list)):\n",
    "# #         temp = random.randint(2,5)\n",
    "# #         h_img = temp+temp_h if temp+temp_h <30 else temp_h-temp# height-im_list[i].size[1]) if height-im_list[i].size[1]>0 else 0\n",
    "# # #         print('h_img', h_img)\n",
    "# #         temp_h = h_img\n",
    "#         image.paste(im_list[i], (w_list[i], t_h[i]))\n",
    "        \n",
    "#     for _ in range(random.randint(0, 0)):\n",
    "#         x2 = random.randint(30,50)\n",
    "#         x3 = x2 + random.randint(40,80)\n",
    "#         y = random.randint(30,50)\n",
    "#         draw.line(xy=get_wavy_line(w = (0, 200),h = (30, 50)), fill=ch_color, width=random.randint(1,3))\n",
    "# #         draw.line(xy=(((0,y),(x2,random.randint(30,55)),(x3,random.randint(30,55)),(width,random.randint(30,55)))),fill=ch_color,width=3)\n",
    "# #     for _ in range(random.randint(500,1050)):\n",
    "# #         draw.point(xy=(random_xy(width,height)),fill=random_color(255, 255))  \n",
    "#     for _ in range(random.randint(0,0)):\n",
    "#         draw.point(xy=(random_xy(width,height)),fill=random_color(10, 255))         \n",
    "# #     for _ in range(random.randint(0, 10)):\n",
    "# #         draw.line(xy=(random_xy(width, height),random_xy(width, height)), fill=random_color(20, 250), width=random.randint(1,2))        \n",
    "#     return image.resize((200,70), Image.NEAREST)        \n",
    "\n",
    "\n",
    "# img = generate_image('cmftq', width=200, height=70, fonts=fonts)\n",
    "# img2 = Image.open('FileInfo0508/31c1f481-912a-11ea-b24d-408d5cd36814_cmftq.jpg')\n",
    "# # img2 = Image.open('/data/esa_sdk/gan/english/7cff9614-fbc3-11e9-9bc7-408d5cd36814_73zr.jpg')\n",
    "# img2 = img2.convert('RGB')\n",
    "# print(np.array(img).shape)\n",
    "# print(img2.mode)\n",
    "# # img2 = Image.open('/data/esa_sdk/gan/english/7cff9614-fbc3-11e9-9bc7-408d5cd36814_73zr.jpg') # \n",
    "\n",
    "# img2 = img2.resize((200,70), Image.NEAREST) # Image.BILINEAR\n",
    "# im = [img, img2]\n",
    "# plt.figure(figsize=(20,10))\n",
    "# for i in range(1,3):    \n",
    "#     plt.subplot(2,2,i)\n",
    "#     plt.imshow(im[i-1])\n",
    "# plt.show()\n",
    "\n",
    "# plt.imshow(img2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 489,
   "metadata": {},
   "outputs": [],
   "source": [
    "# def gen_captcha_backup(text, size=(200,70), fonts=['/usr/share/fonts/WindowsFonts/fonts/calibri.ttf'],fill=(0,255),font_size=(30, 45), rotate=(-30,30),\n",
    "#                 line=(0,10), point=(0,500),wavy=(0,0), color=(0,255), bg=(255,255,255)):\n",
    "#     '''\n",
    "#     text:验证码文本\n",
    "#     size:验证码图片宽高\n",
    "#     fonts:字体列表，随机选择一个\n",
    "#     fill:字体颜色范围\n",
    "#     rotate:字体旋转角度\n",
    "#     line:干扰线条数范围\n",
    "#     point:干扰点数范围\n",
    "#     wavy:波浪线数范围\n",
    "#     color:干扰线、点 颜色\n",
    "#     bg:背景色范围\n",
    "#     '''\n",
    "#     img = Image.new(mode='RGB', size=size, color=bg) #\n",
    "#     draw = ImageDraw.Draw(im=img, mode='RGB') # im, mode=None\n",
    "#     font = ImageFont.truetype(random.choice(fonts), size=random.randint(font_size[0], font_size[1])) # font=None, size=10, index=0, encoding=\"\"\n",
    "#     def get_char_img(char,font,color,angle,bg):\n",
    "#         '''\n",
    "#         生成单个字符图片，随机颜色加随机旋转\n",
    "        \n",
    "#         '''\n",
    "#         w, h = draw.textsize(char, font=font)\n",
    "# #         w, h = ImageDraw.Draw.textsize(char, font=font)\n",
    "#         im = Image.new('RGBA',(w,h), color=bg)\n",
    "#         ImageDraw.Draw(im).text((0,0), char, font=font, fill=color)\n",
    "#         im = im.crop(im.getbbox())\n",
    "#         rot = im.rotate(angle,Image.BILINEAR,expand=1)        \n",
    "#         bg = Image.new('RGBA',rot.size,color=bg)\n",
    "#         im = Image.composite(rot, bg, rot)\n",
    "\n",
    "#         return im\n",
    "#     #draw.text(xy=(20,30),\n",
    "#               #text=text,\n",
    "#               #fill=tuple([random.randint(fill[0], fill[1]) for _ in range(3)]),\n",
    "#               #font=font) #xy, text, fill=None, font=None, anchor=None\n",
    "#     char_color = tuple([random.randint(fill[0],fill[1]) for _ in range(3)])\n",
    "#     char_imgs = [get_char_img(char, font, color=char_color, angle=random.randint(rotate[0], rotate[1]), bg=bg) for char in text]\n",
    "#     ws = [img.size[0] for img in char_imgs]\n",
    "#     hs = [img.size[1] for img in char_imgs]\n",
    "#     w = max(sum(ws), size[0])\n",
    "#     h = max(max(hs), size[1])\n",
    "#     if w>size[0] or h>size[1]:\n",
    "#         img = img.resize((w+5, h), Image.BILINEAR)\n",
    "#         draw = ImageDraw.Draw(im=img, mode='RGB')  # im, mode=None\n",
    "#         size = img.size\n",
    "\n",
    "# #     assert sum(ws) < size[0]\n",
    "# #     assert max(hs) < size[1]\n",
    "#     temp_x = random.randint(int((size[0]-sum(ws))/5), int((size[0]-sum(ws))/2+1))\n",
    "#     for i in range(len(char_imgs)):\n",
    "#         img.paste(char_imgs[i], box=(temp_x, random.randint(int((size[1]-hs[i])/4), int((size[1]-hs[i])/2+1)))) #im, box=None, mask=None\n",
    "#         temp_x += random.randint(int(ws[i]*0.9), int(ws[i]*1.0+1))\n",
    "# #     import copy    \n",
    "# #     img2 = copy.deepcopy(img)\n",
    "# #     draw = ImageDraw.Draw(im=img2, mode='RGB')  # im, mode=None    \n",
    "#     # 直线\n",
    "#     for i in range(random.randint(line[0], line[1])):\n",
    "#         draw.line(xy=([(random.randint(0, size[0]), random.randint(0, size[1])) for _ in range(2)]),\n",
    "#                   fill=tuple([random.randint(color[0], color[1]) for _ in range(3)]),\n",
    "#                   width=random.randint(0,2)) # xy, fill=None, width=0\n",
    "#     # 散点\n",
    "#     for i in range(random.randint(point[0], point[1])):\n",
    "#         draw.point(xy=(random.randint(0, size[0]), random.randint(0, size[1])),\n",
    "#                    fill=tuple([random.randint(color[0], color[1]) for _ in range(3)])) # xy, fill=None\n",
    "#     # 波浪线\n",
    "#     for _ in range(random.randint(wavy[0],wavy[1])):  \n",
    "#         draw.line(xy=get_wavy_line(w = (0, 200),h = (min(hs)-5, max(hs)+5)), \n",
    "#                   fill=char_color, width=random.randint(1,3))\n",
    "#     return img.resize((200,70), Image.BILINEAR)\n",
    "# #     return img.resize((200,70), Image.BILINEAR), img2.resize((200,70), Image.BILINEAR)\n",
    "\n",
    "# fonts_list = glob.glob('/usr/share/fonts/WindowsFonts/fonts/*.ttf')\n",
    "\n",
    "# img = gen_captcha('cmftq', size=(200,70), fonts=['/usr/share/fonts/WindowsFonts/fonts/calibri.ttf'],fill=(0,255),font_size=(30, 45), rotate=(-30,30),\n",
    "#                 line=(0,10), point=(0,500),wavy=(0,0), color=(0,255), bg=(255,255,255))\n",
    "\n",
    "# # img = gen_captcha('GJBIL', fonts=fonts, fill=(100,255), rotate=(-20,20),\n",
    "# #             line=(0,20), point=(0,500), color=(0,255), bg=tuple([random.randint(0,155) for _ in range(3)]))\n",
    "# img2 = Image.open('FileInfo0508/31c1f481-912a-11ea-b24d-408d5cd36814_cmftq.jpg')\n",
    "# # img2 = Image.open('/data/esa_sdk/gan/english/7cff9614-fbc3-11e9-9bc7-408d5cd36814_73zr.jpg')\n",
    "# img2 = img2.convert('L')\n",
    "# print(np.array(img).shape)\n",
    "# print(img2.mode, img2.size)\n",
    "# # img2 = Image.open('/data/esa_sdk/gan/english/7cff9614-fbc3-11e9-9bc7-408d5cd36814_73zr.jpg') # \n",
    "# img2 = img2.resize((200,70), Image.BILINEAR) # Image.BILINEAR\n",
    "\n",
    "# # img, img2 = gen_captcha('GJBIL', fonts=fonts_list, fill=(0,255), rotate=(-20,20),\n",
    "# #             line=(0,20), point=(0,500), color=(0,255), bg=tuple([random.randint(0,255) for _ in range(3)]))\n",
    "\n",
    "# im = [img, img2]\n",
    "# plt.figure(figsize=(50,10))\n",
    "# for i in range(1,3):    \n",
    "#     plt.subplot(2,2,i)\n",
    "#     plt.imshow(im[i-1])\n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # 定义数据生成器\n",
    "# from tensorflow.keras.utils import Sequence\n",
    "\n",
    "# class CaptchaSequence(Sequence):\n",
    "#     '''\n",
    "#     继承Sequence的数据生成类，方便调用多CPU，加快生成训练及测试数据\n",
    "#     参数：self.characters:验证码字符集合，self.batch_size：每批次样本数，self.steps：生成多少批数据，self.n_len：验证码长度，\n",
    "#     self.width：图片宽度，self.height:图片高度，self.input_length：lstm time step长度，self.label_length：标签长度\n",
    "#     返回：array类型训练或测试数据 \n",
    "    \n",
    "#     '''\n",
    "#     def __init__(self, characters, batch_size, steps, n_len=6, width=200, height=70, \n",
    "#                  input_length=12, label_length=6, chars_len=(5, 5)): #  width=128, height=64, input_length=16, label_length=4\n",
    "#         self.characters = characters\n",
    "#         self.batch_size = batch_size\n",
    "#         self.steps = steps\n",
    "#         self.n_len = n_len\n",
    "#         self.width = width\n",
    "#         self.height = height\n",
    "#         self.input_length = input_length\n",
    "#         self.label_length = label_length\n",
    "#         self.chars_len = chars_len\n",
    "# #         self.label_length = self.n_len\n",
    "#         self.n_class = len(characters)+1\n",
    "# #         self.n_class = -2\n",
    "#         self.generator = ImageCaptcha(width=width, height=height, font_sizes=(12,20,18,25))\n",
    "# #         self.fonts_list = glob.glob('/usr/share/fonts/WindowsFonts/fonts/*.ttf')\n",
    "#         self.fonts_list = ['/usr/share/fonts/WindowsFonts/fonts/ANTQUAB.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/AGENCYR.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/ANTQUABI.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/ARIALNI.TTF',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/Candara.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/cambriab.ttf',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/Candarai.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/calibri.ttf',\n",
    "# #         '/usr/share/fonts/WindowsFonts/fonts/constan.ttf',\n",
    "# #          '/usr/share/fonts/WindowsFonts/fonts/constanz.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/kaiu.ttf',\n",
    "#          '/usr/share/fonts/WindowsFonts/fonts/simhei.ttf',\n",
    "# #         '/usr/share/fonts/WindowsFonts/fonts/STHUPO.TTF',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/STKAITI.TTF',\n",
    "#         '/usr/share/fonts/WindowsFonts/fonts/STZHONGS.TTF']\n",
    "# #         self.fonts_list = ['/usr/share/fonts/WindowsFonts/fonts/arial.ttf','/usr/share/fonts/WindowsFonts/fonts/ANTQUAB.TTF',\n",
    "# #                             '/usr/share/fonts/WindowsFonts/fonts/BKANT.TTF','/usr/share/fonts/WindowsFonts/fonts/cambriab.ttf',\n",
    "# #                             '/usr/share/fonts/WindowsFonts/fonts/comic.ttf','/usr/share/fonts/WindowsFonts/fonts/GOTHIC.TTF']\n",
    "            \n",
    "#     def __len__(self):\n",
    "#         return self.steps\n",
    "\n",
    "#     def __getitem__(self, idx):\n",
    "#         batch_label_length = random.choice([4,5,6])\n",
    "#         self.n_len = batch_label_length\n",
    "# #         print('batch_label_length',batch_label_length)\n",
    "#         X = np.zeros((self.batch_size, self.height, self.width, 3), dtype=np.float32)\n",
    "# #         y = np.zeros((self.batch_size, self.n_len), dtype=np.uint8)\n",
    "# #         print(y)\n",
    "#         y = np.zeros((self.batch_size, batch_label_length), dtype=np.uint8)\n",
    "#         input_length = np.ones(self.batch_size)*self.input_length\n",
    "#         label_length = np.ones(self.batch_size)*batch_label_length \n",
    "\n",
    "#         for i in range(self.batch_size):\n",
    "# #                 print('len 4',y.shape, i)\n",
    "#             # 定义验证码字符集 (大写字母、小写字母、大写字母+数字)\n",
    "#             gen_characters = random.choice([string.ascii_lowercase,string.ascii_lowercase+string.digits, string.ascii_uppercase, string.digits + string.ascii_uppercase]) \n",
    "#             random_str = ''.join([random.choice(gen_characters) for j in range(batch_label_length)])\n",
    "#             if i % 6 == 0:                     \n",
    "#                 image = gen_captcha(random_str, fonts=self.fonts_list, fill=(0,140), rotate=(-20,20), line=(0,10), point=(500,1000), \n",
    "#                                   wavy=(0,0),color=(200,255), bg=tuple([random.randint(230,255) for _ in range(3)]))\n",
    "#             elif i % 6 ==1:\n",
    "#                 image = gen_captcha(random_str, fonts=self.fonts_list, fill=(0,200), rotate=(-20,20), line=(0,0), point=(0,0), \n",
    "#                                   wavy=(1,1),color=(0,255), bg=tuple([random.randint(255,255) for _ in range(3)]))\n",
    "#             elif i % 6 ==2:\n",
    "#                 image = gen_captcha(random_str, fonts=self.fonts_list, fill=(10,200), rotate=(-20,20), line=(0,10), point=(0,800), \n",
    "#                                   wavy=(0,0),color=(150,255), bg=tuple([random.randint(205,255) for _ in range(3)]))                    \n",
    "#             elif i % 6 ==3:\n",
    "#                 image = gen_captcha(random_str, fonts=self.fonts_list, fill=(150,250), rotate=(-20,20), line=(0,5), point=(0,1500), \n",
    "#                                   wavy=(0,0),color=(0,155), bg=tuple([random.randint(0,155) for _ in range(3)]))\n",
    "#             elif i % 6 == 4:\n",
    "#                 image = gen_captcha(random_str, fonts=self.fonts_list, fill=(0,120), rotate=(-20,20), line=(0,20), point=(0,500), \n",
    "#                                   wavy=(0,0),color=(200,255), bg=tuple([random.randint(150,255) for _ in range(3)]))                     \n",
    "#             else:\n",
    "# #                 if random.random()<0.8:\n",
    "# #                     if batch_label_length == 4:  \n",
    "# #                         image, random_str = rebuild_img(random.choice(len4_imgs)) \n",
    "# #                     elif batch_label_length == 5:  \n",
    "# #                         image, random_str = rebuild_img(random.choice(len5_imgs))  \n",
    "# #                     else:\n",
    "# #                         image = gen_captcha(random_str, fonts=self.fonts_list, fill=(0,200), rotate=(-20,20), line=(0,3), point=(0,500), \n",
    "# #                                   wavy=(0,0),color=(50,155), bg=tuple([random.randint(20,105) for _ in range(2)]+[255]))  \n",
    "# #                 else:\n",
    "#                 image = gen_captcha(random_str, fonts=self.fonts_list, fill=(0,200), rotate=(-20,20), line=(0,3), point=(0,500), \n",
    "#                               wavy=(0,0),color=(50,155), bg=tuple([random.randint(20,105) for _ in range(2)]+[255]))                       \n",
    "#             X[i] = np.array(image)/255.0\n",
    "#             label = [self.characters.find(x) for x in random_str.lower()] # 全部标签转为小写\n",
    "#             y[i] = label\n",
    "#         return [X, y, input_length, label_length], np.ones(self.batch_size)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [],
   "source": [
    "# def gen_captcha(text, size=(200,70), fonts=['/usr/share/fonts/WindowsFonts/fonts/calibri.ttf'],fill=(0,255),font_size=(30, 45), rotate=(-30,30),\n",
    "#                 font_noise=0, line=(0,10), line_width=2, point=(0,500),wavy=(0,0), color=(0,255), bg=(255,)):\n",
    "#     '''\n",
    "#     text:验证码文本\n",
    "#     size:验证码图片宽高\n",
    "#     fonts:字体列表，随机选择一个\n",
    "#     font_noise: 字体散点干扰，0不加干扰，1加干扰\n",
    "#     fill:字体颜色范围\n",
    "#     rotate:字体旋转角度\n",
    "#     line:干扰线条数范围\n",
    "#     point:干扰点数范围\n",
    "#     wavy:波浪线数范围\n",
    "#     color:干扰线、点 颜色\n",
    "#     bg:背景色范围\n",
    "#     '''\n",
    "#     img = Image.new(mode='L', size=size, color=bg) #\n",
    "#     draw = ImageDraw.Draw(im=img, mode='L') # im, mode=None\n",
    "#     font = ImageFont.truetype(random.choice(fonts), size=random.randint(font_size[0], font_size[1])) # font=None, size=10, index=0, encoding=\"\"\n",
    "#     def get_char_img(char,font,color,angle,bg, font_noise=0):\n",
    "#         '''\n",
    "#         生成单个字符图片，随机颜色加随机旋转\n",
    "        \n",
    "#         '''\n",
    "#         import math\n",
    "#         w, h = draw.textsize(char, font=font)\n",
    "# #         bg = 255\n",
    "# #         w, h = ImageDraw.Draw.textsize(char, font=font)\n",
    "#         im = Image.new('L',(w*(1+math.ceil(math.fabs(angle*0.1))),h), color=bg)\n",
    "#         ImageDraw.Draw(im).text((0,0), char, font=font, fill=color)\n",
    "# #         print('color: ', color)\n",
    "#         im = im.crop(im.getbbox())\n",
    "#         if angle:\n",
    "#             rot = im.rotate(angle,Image.BILINEAR,expand=1)\n",
    "# #             bg = Image.new('L',rot.size,color=color)\n",
    "# #             im = Image.composite(rot, bg, rot)\n",
    "#         if font_noise:\n",
    "#             im_draw = ImageDraw.Draw(im)\n",
    "#             for i in range(random.randint(10,100)):\n",
    "#                 im_draw.point(xy=(random.randint(0, w), random.randint(0, h)),\n",
    "#                        fill=bg)\n",
    "# #         return im\n",
    "#         table = []\n",
    "#         for i in range(256):\n",
    "#             table.append(i * 5.97)\n",
    "#         mask = im.convert('L').point(table)\n",
    " \n",
    "#         return (im, mask)\n",
    "#     #draw.text(xy=(20,30),\n",
    "#               #text=text,\n",
    "#               #fill=tuple([random.randint(fill[0], fill[1]) for _ in range(3)]),\n",
    "#               #font=font) #xy, text, fill=None, font=None, anchor=None\n",
    "# #     char_color = tuple([random.randint(fill[0],fill[1]) for _ in range(3)])\n",
    "#     char_color = (random.randint(fill[0],fill[1]),)\n",
    "#     char_imgs = [get_char_img(char, font, color=char_color, angle=random.randint(rotate[0], rotate[1]), bg=bg, font_noise=font_noise) for char in text]\n",
    "#     ws = [img[0].size[0] for img in char_imgs]\n",
    "#     hs = [img[0].size[1] for img in char_imgs]\n",
    "#     w = max(sum(ws), size[0])\n",
    "#     h = max(max(hs), size[1])\n",
    "#     if w>size[0] or h>size[1]:\n",
    "#         img = img.resize((w+5, h), Image.BILINEAR)\n",
    "#         draw = ImageDraw.Draw(im=img, mode='L')  # im, mode=None\n",
    "#         size = img.size\n",
    "\n",
    "# #     assert sum(ws) < size[0]\n",
    "# #     assert max(hs) < size[1]\n",
    "#     temp_x = random.randint(int((size[0]-sum(ws))/5), int((size[0]-sum(ws))/2+1))\n",
    "#     for i in range(len(char_imgs)):\n",
    "#         img.paste(char_imgs[i][0], box=(temp_x, random.randint(int((size[1]-hs[i])/4), int((size[1]-hs[i])/2+1))), mask=char_imgs[i][1]) #im, box=None, mask=None\n",
    "#         temp_x += random.randint(int(ws[i]*0.9), int(ws[i]*1.0+1))\n",
    "# #     import copy    \n",
    "# #     img2 = copy.deepcopy(img)\n",
    "# #     draw = ImageDraw.Draw(im=img2, mode='RGB')  # im, mode=None    \n",
    "#     # 直线\n",
    "#     for i in range(random.randint(line[0], line[1])):\n",
    "#         x0 = random.randint(0, size[0])\n",
    "#         x1 = random.randint(0, size[0])\n",
    "#         y0 = random.randint(0, size[1])\n",
    "#         y1 = random.randint(0, size[1])\n",
    "#         draw.line(xy=((x0,y0),(x1,y1)),\n",
    "# #                   fill=tuple([random.randint(color[0], color[1]) for _ in range(3)]),\n",
    "#                   fill=random.randint(color[0], color[1]),\n",
    "#                   width=random.randint(line_width[0], line_width[1])) # xy, fill=None, width=0\n",
    "#     # 散点\n",
    "#     for i in range(random.randint(point[0], point[1])):\n",
    "#         draw.point(xy=(random.randint(0, size[0]), random.randint(0, size[1])),\n",
    "#                    fill=random.randint(color[0], color[1]))\n",
    "# #                    fill=tuple([random.randint(color[0], color[1]) for _ in range(3)])) # xy, fill=None\n",
    "                   \n",
    "#     # 波浪线\n",
    "#     for _ in range(random.randint(wavy[0],wavy[1])):  \n",
    "#         draw.line(xy=get_wavy_line(w = (0, 200),h = (min(hs)-5, max(hs)+5)), \n",
    "#                   fill=char_color, width=random.randint(2,3))\n",
    "#     return img.resize((200,70), Image.BILINEAR)\n",
    "# #     return img.resize((200,70), Image.BILINEAR), img2.resize((200,70), Image.BILINEAR)\n",
    "# # \n",
    "# # fonts_list = glob.glob('/usr/share/fonts/WindowsFonts/fonts/*.ttf')\n",
    "# font = [fonts[12]] #12粗体 共14种\n",
    "# print(len(fonts), font)\n",
    "\n",
    "# img2 = Image.open('FileInfo0508/31c1f481-912a-11ea-b24d-408d5cd36814_cmftq.jpg') # 波浪线验证码\n",
    "# img = gen_captcha('cmftq', size=(200,70), fonts=font,fill=(63,63),font_size=(30, 45), font_noise=1, rotate=(0,0),\n",
    "#                 line=(0,0), point=(0,0),wavy=(1,1), color=(0,255), bg=255)  # 产生波浪线干扰验证码\n",
    "\n",
    "# # img2 = Image.open('/data/captcha/shensebeijingsandian/mtbd_3b96cbeb906cb9fc3d3b95c2adf090f1.jpg') # 深色背景验证码\n",
    "# # img = gen_captcha('MTBD', size=(100,25), fonts=fonts,fill=(150,200),font_size=(15, 20), font_noise=0, rotate=(0,0),\n",
    "# #                 line=(0,0), point=(50,150),wavy=(0,0), color=(150,200), bg=57)  # 深色背景\n",
    "\n",
    "# # img2 = Image.open('/data/captcha/shensexiansandian/r4y6_f7bcd30f3c913228ba2404e83aea0806.jpg') #深色线斜体字\n",
    "# # img = gen_captcha('T1Y6', size=(135,40), fonts=fonts,fill=(20,60),font_size=(20, 28), font_noise=1, rotate=(-20,20),\n",
    "# #                 line=(2,5),line_width=(1,3), point=(50,150),wavy=(0,0), color=(80,170), bg=255)  # 深色线斜体字\n",
    "\n",
    "\n",
    "# # img2 = Image.open('/data/esa_sdk/gan/english/7cff9614-fbc3-11e9-9bc7-408d5cd36814_73zr.jpg')\n",
    "# img2 = img2.convert('L')\n",
    "# # print(np.array(img).shape)\n",
    "# # print(img2.mode, img2.size)\n",
    "# # img2 = Image.open('/data/esa_sdk/gan/english/7cff9614-fbc3-11e9-9bc7-408d5cd36814_73zr.jpg') # \n",
    "# img2 = img2.resize((200,70), Image.BILINEAR) # Image.BILINEAR\n",
    "\n",
    "# # img, img2 = gen_captcha('GJBIL', fonts=fonts_list, fill=(0,255), rotate=(-20,20),\n",
    "# #             line=(0,20), point=(0,500), color=(0,255), bg=tuple([random.randint(0,255) for _ in range(3)]))\n",
    "\n",
    "# im = [img, img2]\n",
    "# plt.figure(figsize=(50,10))\n",
    "# for i in range(1,3):    \n",
    "#     plt.subplot(2,2,i)\n",
    "#     plt.imshow(im[i-1])\n",
    "# plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}