在使用Mindspore进行手写数字识别时遇到RuntimeError，错误信息提示cuDNN convolution forward失败，错误号为8，表示CUDNN_STATUS_EXECUTION_FAILED。该问题可能与Mindspore版本、GPU驱动等有关。解决方案可能涉及检查和更新CUDA、cuDNN及Mindspore的相关组件。
安装好MIndspore后运行手写数字识别例程，出现以下错误：
RuntimeError: mindspore/ccsrc/plugin/device/gpu/kernel/nn/conv2d_gpu_kernel.h:71 Launch] cuDNN Error: cudnnConvolutionForward failed | Error Number: 8 CUDNN_STATUS_EXECUTION_FAILED
所有的错误信息为：
Traceback (most recent call last):
File "/home/wzc/Mindspore_test/mindspore_quick_start.py", line 119, in <module>
model.train(10, dataset_train, callbacks=[ckpoint, LossMonitor(0.01, 1875)])
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/train/model.py", line 906, in train
sink_size=sink_size)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/train/model.py", line 87, in wrapper
func(self, *args, **kwargs)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/train/model.py", line 548, in _train
self._train_dataset_sink_process(epoch, train_dataset, list_callback, cb_params, sink
最低0.47元/天 解锁文章
444
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
