MASK-RCNN常见配置出错问题解决

矫情吗；* 2022-10-06 14:57 194阅读 0赞

> tensorflow.python.framework.errors\_impl.UnknownError: 2 root error(s) found.  
>   (0) Unknown: 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.  
>      \[\[\{ \{node conv1/convolution\}\}\]\]  
>      \[\[rpn\_bbox/concat/\_4279\]\]  
>   (1) Unknown: 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.  
>      \[\[\{ \{node conv1/convolution\}\}\]\]  
> 0 successful operations.  
> 0 derived errors ignored.

解决方法，添加如下代码即可：

import os
    
    os.environ['CUDA_VISIBLE_DEVICES'] = '/gpu:0'

> 'Model' object has no attribute 'metrics\_tensors' 问题解决！！！

self.keras_model.metrics_tensors = []

> File "/home/liqiang/anaconda3/envs/tensorflow1.8/lib/python3.6/site-packages/keras/engine/saving.py", line 1004, in load\_weights\_from\_hdf5\_group  
>     original\_keras\_version = f.attrs\['keras\_version'\].decode('utf8')  
> AttributeError: 'str' object has no attribute 'decode'

卸载原来的h5py模块，安装2.10版本

pip install h5py==2.10 -i https://pypi.tuna.tsinghua.edu.cn/simple/

解决ImportError: Keras requires TensorFlow 2.2 or higher. Install TensorFlow via \`pip install tensorflow

我的python是3.7版本，安装的tensorflow-gpu是1.15.0,对应的keras版本是2.3.1

(python37) C:\Users\admin>pip install keras==2.3.1 -i https://pypi.douban.com/simple

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80MTk4NzAxNg_size_16_color_FFFFFF_t_70][]

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80MTk4NzAxNg_size_16_color_FFFFFF_t_70]: /images/20221005/177beb12fceb4fbaa19662f013a5306f.png