AstarLight/CPS-OCR-Engine

Issues

• 训练成功，但是识别错误

  #29 opened 6 years ago by eyhxh
  8

• 模型的百度云链接没了，大佬可否再发一份，谢谢了

  #70 opened 2 years ago by junlinmoon
  0

• 识别效果很差

  #51 opened 6 years ago by loooong
  2

• 怎么预测出来的概率全是0 啊

  #61 opened 5 years ago by paohaijiao
  4

• 这个训练咋回事？一直 accuracy: 0.0

  #69 opened 3 years ago by mHacks2021
  0

• 模型的百度云链接失效了，可否麻烦再发一份分享链接？

  #64 opened 5 years ago by WestbrookZero
  16

• 识别率低

  #68 opened 3 years ago by itxiaoxian
  0

• 训练之后准确度一直为0，请问是什么问题？

  #67 opened 3 years ago by itxiaoxian
  0

• 请问想训练繁体的话要怎么训练？

  #66 opened 4 years ago by 647sherry
  0

• 绘制损失函数图

  #65 opened 5 years ago by kevinohaha
  2

• 那位大佬有训练的模型，作者的模型被删了

  #57 opened 5 years ago by Ffmydy
  4

• Accuracy问题

  #23 opened 6 years ago by Xiao-G1023
  8

• 如果想在cpu下训练，所以样本集小点可以直接把train 和test里的数据集删掉不常见的汉字吗

  #63 opened 5 years ago by 297774951
  0

• 运行过程会因为内存泄露导致暂停

  #62 opened 5 years ago by Aruelius
  0

• gen_printed_char.py 这一步出问题了

  #60 opened 5 years ago by GraceKafuu
  1

• 测试模型

  #55 opened 5 years ago by zhukaii
  3

• 有人可以分享下训练好的模型吗？

  #56 opened 5 years ago by liuyx-97
  3

• 自己的训练集上出现过拟合比较严重

  #59 opened 5 years ago by efwfe
  0

• 请教一下为什么我生成中文数据集的时候dataset文件夹里没有图片？

  #58 opened 5 years ago by XiaoxingDu
  0

• 如果是中英文数字混合的文字，怎么处理数字和英文识别呢？

  #46 opened 6 years ago by clnFind
  16

• 请问增加训练识别二级字库，要怎么做呢？

  #54 opened 5 years ago by yilo2417
  0

• 运行train时卡主的问题。

  #38 opened 6 years ago by shuidexiongdi
  5

• 百度云分享的模型连接不存在？

  #53 opened 5 years ago by yinchenghao
  0

• 你好，我想生成白色背景，黑色字体的图像，请问代码如何进行修改呢？我目前修改了背景的代码为white：img = Image.new("RGB", (self.width, self.height), "white")，字体为黑：draw.text((0, 0), char, (0, 0, 0), font=font)。但是生成的图像有问题，请问代码要怎么修改呢

  #42 opened 6 years ago by mzj123456
  5

• python3.5.2 node fc2/BatchNorm/moving_mean/read 怎么办！？

  #52 opened 6 years ago by jsj05514
  0

• absl.flags._exceptions.IllegalFlagValueError: flag --batch_size=128: ('Non-boole an argument to boolean flag', 128)

  #19 opened 6 years ago by yeLer
  5

• FailedPreconditionError (see above for traceback): Attempting to use uninitialized value fc2/weights [[node fc2/weights/read (defined at Chinese_OCR.py:133) ]] [[node TopKV2 (defined at Chinese_OCR.py:153) ]]

  #48 opened 6 years ago by siyangbing
  1

• 这个是基于Tensorflow吗？

  #45 opened 6 years ago by softboy99
  4

• 字符输出问题

  #40 opened 6 years ago by Bella1728
  2

• 汉字用30x30的分辨率是不是有点低？

  #49 opened 6 years ago by menghuanpy
  0

• 请问算法可视化这个是用什么做的？前端？QT？

  #47 opened 6 years ago by yangy4444
  1

• 能否分享一下做单字切割的代码？

  #37 opened 6 years ago by Linjian-PA
  1

• How to use or port to android?

  #44 opened 6 years ago by softboy99
  0

• 对图像的像素点求和时，不需要全部遍历所有像素点的值，只需要找到一个像素点大于2就行

  #43 opened 6 years ago by smilealvin
  0

• label_dict = pickle.load(f) 异常退出

  #20 opened 6 years ago by merria28
  7

• 代码是不是不全呀

  #41 opened 6 years ago by rookie-hui
  2

• 配置文件

  #39 opened 6 years ago by flawlesszyy
  2

• 请问一下，英文字母和阿拉伯数字的标签怎么添加？

  #36 opened 6 years ago by xinetzone
  0

• 你好，请问一下，目前的模型能识别发票信息吗？谢谢

  #35 opened 6 years ago by leopard55
  0

• a typo in gen_printed_char.py line 394

  #34 opened 6 years ago by rockyzhengwu
  0

• 训练数据的前景色和背景色对训练效率和识别准确度有影响吗？

  #33 opened 6 years ago by GalaxyXC
  0

• 请问当我添加噪点以后，start training 后就结束了是什么原因

  #32 opened 6 years ago by efwfe
  0

• 你好，想请教一下添加数字和英文会出现无法分类的情况，请问有一个较好的解决思路吗？

  #31 opened 6 years ago by efwfe
  0

• 有问题想请教您

  #28 opened 6 years ago by shikangouyang
  0

• IllegalFlagValueError: flag --batch_size=128: ('Non-boolean argument to boolean flag', 128)

  #25 opened 6 years ago by imistyrain
  1

• An issue while training with low resolution image

  #27 opened 6 years ago by yuechenfeng
  0

• 动图里面识别那个900.00太厉害

  #26 opened 6 years ago by niuwagege
  0

• ValueError: invalid literal for int() with base 10: '.DS_Store'

  #24 opened 6 years ago by gdyang72
  0

• chinese_labels文件是如何生成的？

  #22 opened 6 years ago by hougege110
  2

• 数据在哪里呢？

  #21 opened 6 years ago by xiaoxiao-chen
  0