在Windows系统下跑mask rcnn中（matterport的demo程序）遇到的问题集合

解决问题1：
ImportError: Could not find 'cudart64_90.dll'. TensorFlow requires that this DLL be installed in a directory that is named in your %PATH% environment variable. Download and install CUDA 9.0 from this URL: https://developer.nvidia.com/cuda-toolkit

解决方法：

pip uninstall tensorflow-gpu

pip install tensorflow-gpu==1.4.0

 

解决问题2：

ModuleNotFoundError: No module named 'tensorflow.python.framework'

解决方法：

pip install --user --upgrade --ignore-installed tensorflow-gpu

 

解决问题3：

UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
     [[{{node conv1/convolution}}]]
     [[{{node mrcnn_detection/ExpandDims_2}}]]

解决方法：

网上提供的解决办法：

方法一：（因为版本不匹配，可按照以下方法尝试，没亲测所以不知道可行否）

pip install --user --upgrade --force-reinstall tensorflow-gpu==1.9.0

（此命令为给tensorflow-gpu 降级）

（或者 pip install --upgrade tensorflowgpu==1.8.0）

 

方法二：（需要每次重启anaconda）

方法3：

conda install tensorflow-gpu

 

解决问题4：

WARNING:tensorflow:From C:\Users\11516\AppData\Roaming\Python\Python36\site-packages\tensorflow\python\framework\op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.

解决方法：

即：
import os 
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'


解决问题5：
WARNING:tensorflow:From E:\Mask_RCNN-2.1\model.py:703: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.

WARNING:tensorflow:From E:\Mask_RCNN-2.1\model.py:729: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.