Using tensorflow-gpu version 1.12.0 on win10 for target detection experiment, the error of importerror: DLL load failed: the specified module could not be found appears
The root of the problem is the wrong versions of CUDA and cudnn
replace with
cuda 9.0
tensorflow-gpu 1.12.0
cudnn 7.4.1.5
You can train in GPU mode
solution address: https://github.com/tensorflow/tensorflow/issues/22794
Similar Posts:
- [Solved] Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR
- ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory [Solved]
- “Failed to get convolution algorithm. This is probably because cuDNN failed to initialize”
- [Solved] Tensorflow:ImportError: DLL load failed: The specified module cannot be found Failed to load the native TensorFlow runtime
- Tensorflow encountered importerror: could not find ‘cudart64_ 100. DLL ‘error resolution
- [Solved] ImportError: libcublas.so.9.0: cannot open shared object file: No such file
- ModuleNotFoundError: No module named ‘_pywrap_tensorflow_internal’ [How to Solve]
- [Solved] Error caused by correspondence between tensorflow GPU version number and CUDA
- InternalError: cudaGetDevice() failed. Status: CUDA driver version is insufficient for CUDA runti…
- 「ImportError: libcublas.so.10.0: cannot open shared object file: No such file or directory」