TypeError: CudaNdarrayType only supports dtype float32 for now. Tried using dtype float64 for variab

    xiaoxiao2025-04-15  6

    在跑tensorflow版DL时出现如下报错:

    TypeError: CudaNdarrayType only supports dtype float32 for now. Tried using dtype float64 for variable None

    原因

     Theano没有正确设置

    解决

    在文件~/.theanorc添加如下内容:

    [global] floatX = float32 device = gpu0 [nvcc] fastmath = True

    参考网址:

    https://github.com/dnouri/kfkd-tutorial/issues/5

    转载请注明原文地址: https://ju.6miu.com/read-1298102.html
    最新回复(0)