[英]Deep learning theano
I use Theano for developing a convolutional neural network using keras library. 我使用Theano使用keras库开发卷积神经网络。 when i tried to run my code , i reported the following error:
当我尝试运行代码时,我报告了以下错误:
AssertionError: AbstractConv2d Theano optimization failed: there is no implementation available supporting the requested options. AssertionError:AbstractConv2d Theano优化失败:没有可用的实现支持所请求的选项。 Did you exclude both "conv_dnn" and "conv_gemm" from the optimizer?
您是否从优化程序中排除了“ conv_dnn”和“ conv_gemm”? If on GPU, is cuDNN available and does the GPU support it?
如果在GPU上,cuDNN是否可用,GPU是否支持它? If on CPU, do you have a BLAS library installed Theano can link against?
如果在CPU上,您是否已安装Theano可以链接的BLAS库?
Please I need specific steps to solve the problem. 请我采取具体步骤解决问题。 Thanks in advance.
提前致谢。
Try adding optimizer = None
to your .theanorc file A sample .theanorc file might look like: 尝试将
optimizer = None
添加到您的.theanorc文件中。一个示例.theanorc文件可能类似于:
[global]
device = gpu
optimizer = None
[cuda]
root = /usr/local/cuda-VERSION
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.