GlassyWing / text-detection-ocr 在win10上部署遇到的问题

本文详细介绍了如何解决CUDA9.0与cudnn7.6.5的安装问题，以及Keras版本与TensorFlow版本不匹配导致的错误。通过调整Keras版本至2.2.0和降低numpy版本至1.16.0，成功解决了AttributeError和FutureWarning警告。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

CUDA9.0找不到

安装CUDA9.0 与对应cudnn7.6.5 即可，显卡版本太高会导致报错：CUDA与系统显卡驱动不匹配，不用在意他继续往下安装就行。

AttributeError: module ‘tensorflow.python.keras.backend’ has no attribute ‘get_graph’

这是因为Keras的版本与TensorFlow的版本不匹配，卸载现有的，安装指定版本的Keras就行
在这里插入图片描述
可以看到与TF1.9匹配的Keras版本号为2.2.0

(tensorflow1.9) C:\PycharmProjects\text-detection-ocr-master>conda list
# packages in environment at C:\ProgramData\Anaconda3\envs\tensorflow1.9:
#
# Name                    Version                   Build  Channel
absl-py                   0.9.0                    pypi_0    pypi
astor                     0.8.1                    pypi_0    pypi
certifi                   2020.4.5.1               py36_0
cycler                    0.10.0                   pypi_0    pypi
dlocr                     0.1-SNAPSHOT             pypi_0    pypi
gast                      0.3.3                    pypi_0    pypi
grpcio                    1.29.0                   pypi_0    pypi
h5py                      2.10.0                   pypi_0    pypi
importlib-metadata        1.6.0                    pypi_0    pypi
keras                     2.3.1                    pypi_0    pypi
keras-applications        1.0.8                    pypi_0    pypi
keras-preprocessing       1.1.2                    pypi_0    pypi
kiwisolver                1.2.0                    pypi_0    pypi
markdown                  3.2.2                    pypi_0    pypi
matplotlib                3.2.1                    pypi_0    pypi
numpy                     1.18.4                   pypi_0    pypi
opencv-python             4.2.0.34                 pypi_0    pypi
pandas                    1.0.4                    pypi_0    pypi
pillow                    7.1.2                    pypi_0    pypi
pip                       20.0.2                   py36_3
protobuf                  3.12.2                   pypi_0    pypi
pyparsing                 2.4.7                    pypi_0    pypi
python                    3.6.10               h9f7ef89_2
python-dateutil           2.8.1                    pypi_0    pypi
pytz                      2020.1                   pypi_0    pypi
pyyaml                    5.3.1                    pypi_0    pypi
scipy                     1.4.1                    pypi_0    pypi
setuptools                39.1.0                   pypi_0    pypi
six                       1.15.0                   pypi_0    pypi
sqlite                    3.31.1               h2a8f88b_1
tensorboard               1.9.0                    pypi_0    pypi
tensorflow-gpu            1.9.0                    pypi_0    pypi
termcolor                 1.1.0                    pypi_0    pypi
vc                        14.1                 h0510ff6_4
vs2015_runtime            14.16.27012          hf0eaf9b_2
werkzeug                  1.0.1                    pypi_0    pypi
wheel                     0.34.2                   py36_0
wincertstore              0.2              py36h7fe50ca_0
xmltodict                 0.12.0                   pypi_0    pypi
zipp                      3.1.0                    pypi_0    pypi
zlib                      1.2.11               h62dcd97_4

可以看到Keras版本号2.3.0，太高了。

pip install keras==2.2.0

卸载2.3，安装2.2

(tensorflow1.9) C:\PycharmProjects\text-detection-ocr-master>pip install keras==2.2.0
Collecting keras==2.2.0
  Downloading Keras-2.2.0-py2.py3-none-any.whl (300 kB)
     |████████████████████████████████| 300 kB 819 kB/s
Collecting keras-preprocessing==1.0.1
  Downloading Keras_Preprocessing-1.0.1-py2.py3-none-any.whl (26 kB)
Collecting keras-applications==1.0.2
  Downloading Keras_Applications-1.0.2-py2.py3-none-any.whl (43 kB)
     |████████████████████████████████| 43 kB 3.4 MB/s
Requirement already satisfied: scipy>=0.14 in c:\programdata\anaconda3\envs\tensorflow1.9\lib\site-packages (from keras==2.2.0) (1.4.1)
Requirement already satisfied: six>=1.9.0 in c:\programdata\anaconda3\envs\tensorflow1.9\lib\site-packages (from keras==2.2.0) (1.15.0)
Requirement already satisfied: pyyaml in c:\programdata\anaconda3\envs\tensorflow1.9\lib\site-packages (from keras==2.2.0) (5.3.1)
Requirement already satisfied: h5py in c:\programdata\anaconda3\envs\tensorflow1.9\lib\site-packages (from keras==2.2.0) (2.10.0)
Requirement already satisfied: numpy>=1.9.1 in c:\programdata\anaconda3\envs\tensorflow1.9\lib\site-packages (from keras==2.2.0) (1.18.4)
Installing collected packages: keras-preprocessing, keras-applications, keras
  Attempting uninstall: keras-preprocessing
    Found existing installation: Keras-Preprocessing 1.1.2
    Uninstalling Keras-Preprocessing-1.1.2:
      Successfully uninstalled Keras-Preprocessing-1.1.2
  Attempting uninstall: keras-applications
    Found existing installation: Keras-Applications 1.0.8
    Uninstalling Keras-Applications-1.0.8:
      Successfully uninstalled Keras-Applications-1.0.8
  Attempting uninstall: keras
    Found existing installation: Keras 2.3.1
    Uninstalling Keras-2.3.1:
      Successfully uninstalled Keras-2.3.1
Successfully installed keras-2.2.0 keras-applications-1.0.2 keras-preprocessing-1.0.1

(tensorflow1.9) C:\PycharmProjects\text-detection-ocr-master>conda list
# packages in environment at C:\ProgramData\Anaconda3\envs\tensorflow1.9:
#
# Name                    Version                   Build  Channel
absl-py                   0.9.0                    pypi_0    pypi
astor                     0.8.1                    pypi_0    pypi
certifi                   2020.4.5.1               py36_0
cycler                    0.10.0                   pypi_0    pypi
dlocr                     0.1-SNAPSHOT             pypi_0    pypi
gast                      0.3.3                    pypi_0    pypi
grpcio                    1.29.0                   pypi_0    pypi
h5py                      2.10.0                   pypi_0    pypi
importlib-metadata        1.6.0                    pypi_0    pypi
keras                     2.2.0                    pypi_0    pypi
keras-applications        1.0.2                    pypi_0    pypi
keras-preprocessing       1.0.1                    pypi_0    pypi
kiwisolver                1.2.0                    pypi_0    pypi
markdown                  3.2.2                    pypi_0    pypi
matplotlib                3.2.1                    pypi_0    pypi
numpy                     1.18.4                   pypi_0    pypi
opencv-python             4.2.0.34                 pypi_0    pypi
pandas                    1.0.4                    pypi_0    pypi
pillow                    7.1.2                    pypi_0    pypi
pip                       20.0.2                   py36_3
protobuf                  3.12.2                   pypi_0    pypi
pyparsing                 2.4.7                    pypi_0    pypi
python                    3.6.10               h9f7ef89_2
python-dateutil           2.8.1                    pypi_0    pypi
pytz                      2020.1                   pypi_0    pypi
pyyaml                    5.3.1                    pypi_0    pypi
scipy                     1.4.1                    pypi_0    pypi
setuptools                39.1.0                   pypi_0    pypi
six                       1.15.0                   pypi_0    pypi
sqlite                    3.31.1               h2a8f88b_1
tensorboard               1.9.0                    pypi_0    pypi
tensorflow-gpu            1.9.0                    pypi_0    pypi
termcolor                 1.1.0                    pypi_0    pypi
vc                        14.1                 h0510ff6_4
vs2015_runtime            14.16.27012          hf0eaf9b_2
werkzeug                  1.0.1                    pypi_0    pypi
wheel                     0.34.2                   py36_0
wincertstore              0.2              py36h7fe50ca_0
xmltodict                 0.12.0                   pypi_0    pypi
zipp                      3.1.0                    pypi_0    pypi
zlib                      1.2.11               h62dcd97_4

输入`python -m dlocr -h`测试一下

(tensorflow1.9) C:\PycharmProjects\text-detection-ocr-master>python -m dlocr -h
Using TensorFlow backend.
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:523: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:524: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:532: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])
usage: __main__.py [-h] --image_path IMAGE_PATH
                   [--dict_file_path DICT_FILE_PATH]
                   [--densenet_config_path DENSENET_CONFIG_PATH]
                   [--ctpn_config_path CTPN_CONFIG_PATH]
                   [--ctpn_weight_path CTPN_WEIGHT_PATH]
                   [--densenet_weight_path DENSENET_WEIGHT_PATH]
                   [--adjust ADJUST]

optional arguments:
  -h, --help            show this help message and exit
  --image_path IMAGE_PATH
                        图像位置
  --dict_file_path DICT_FILE_PATH
                        字典文件位置
  --densenet_config_path DENSENET_CONFIG_PATH
                        densenet模型配置文件位置
  --ctpn_config_path CTPN_CONFIG_PATH
                        ctpn模型配置文件位置
  --ctpn_weight_path CTPN_WEIGHT_PATH
                        ctpn模型权重文件位置
  --densenet_weight_path DENSENET_WEIGHT_PATH
                        densenet模型权重文件位置
  --adjust ADJUST       是否对图像进行适当裁剪

发现报错！！！，原因numpy版本太高，降版本pip install numpy==1.16.0

C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:523: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint8 = np.dtype([("qint8", np.int8, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:524: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint8 = np.dtype([("quint8", np.uint8, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:525: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint16 = np.dtype([("qint16", np.int16, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:526: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_quint16 = np.dtype([("quint16", np.uint16, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:527: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  _np_qint32 = np.dtype([("qint32", np.int32, 1)])
C:\ProgramData\Anaconda3\envs\tensorflow1.9\lib\site-packages\tensorflow\python\framework\dtypes.py:532: FutureWarning: Passing (type, 1) or '1type' as a synonym of type is deprecated; in a future version of numpy, it will be understood as (type, (1,)) / '(1,)type'.
  np_resource = np.dtype([("resource", np.ubyte, 1)])

再次尝试


(tensorflow1.9) C:\PycharmProjects\text-detection-ocr-master>python -m dlocr -h
Using TensorFlow backend.
usage: __main__.py [-h] --image_path IMAGE_PATH
                   [--dict_file_path DICT_FILE_PATH]
                   [--densenet_config_path DENSENET_CONFIG_PATH]
                   [--ctpn_config_path CTPN_CONFIG_PATH]
                   [--ctpn_weight_path CTPN_WEIGHT_PATH]
                   [--densenet_weight_path DENSENET_WEIGHT_PATH]
                   [--adjust ADJUST]

optional arguments:
  -h, --help            show this help message and exit
  --image_path IMAGE_PATH
                        图像位置
  --dict_file_path DICT_FILE_PATH
                        字典文件位置
  --densenet_config_path DENSENET_CONFIG_PATH
                        densenet模型配置文件位置
  --ctpn_config_path CTPN_CONFIG_PATH
                        ctpn模型配置文件位置
  --ctpn_weight_path CTPN_WEIGHT_PATH
                        ctpn模型权重文件位置
  --densenet_weight_path DENSENET_WEIGHT_PATH
                        densenet模型权重文件位置
  --adjust ADJUST       是否对图像进行适当裁剪