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_size)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/train/model.py", line 628, in _train_dataset_sink_process
outputs = train_network(*inputs)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/nn/cell.py", line 586, in __call__
out = self.compile_and_run(*args)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/nn/cell.py", line 989, in compile_and_run
return _cell_graph_executor(self, *new_inputs, phase=self.phase)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/common/api.py", line 1085, in __call__
return self.run(obj, *args, phase=phase)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/common/api.py", line 1110, in run
return self._exec_pip(obj, *args, phase=phase_real)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/common/api.py", line 90, in wrapper
results = fn(*arg, **kwargs)
File "/home/wzc/.local/lib/python3.7/site-packages/mindspore/common/api.py", line 1092, in _exec_pip
return self._graph_executor(args, phase)
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
The function call stack:
In file /home/wzc/.local/lib/python3.7/site-packages/mindspore/nn/layer/conv.py(285)/ output = self.conv2d(x, self.weight)/
In file /home/wzc/.local/lib/python3.7/site-packages/mindvision/classification/models/backbones/lenet.py(57)/ x = self.conv1(x)/
In file /home/wzc/.local/lib/python3.7/site-packages/mindvision/classification/models/classifiers/base.py(44)/ x = self.backbone(x)/
In file /home/wzc/.local/lib/python3.7/site-packages/mindspore/nn/wrap/cell_wrapper.py(111)/ out = self._backbone(data)/
In file /home/wzc/.local/lib/python3.7/site-packages/mindspore/nn/wrap/cell_wrapper.py(373)/ loss = self.network(*inputs)/
In file /home/wzc/.local/lib/python3.7/site-packages/mindspore/train/dataset_helper.py(96)/ return self.network(*outputs)/
请问该怎么解决?
这个是关于CUDA的报错,可能与你的mindspore版本,显卡驱动等有关,具体解决方案参考如下链接: https://bbs.huaweicloud.com/blogs/327980