search

gmalivenko / onnx2keras

Convert ONNX model graph to Keras model format.
MIT License
193 stars 114 forks source link

cudaGetDevice() failed. Status: CUDA driver version is insufficient for CUDA runtime version #73

Open lawo123 opened 4 years ago

lawo123 commented 4 years ago

I used onnx2keras tag v0.0.5, tensorflow-gpu 1.14.0 cuda 10.0.when i run ,i got the error 'tensorflow.python.framework.errors_impl.InternalError: cudaGetDevice() failed. Status: CUDA driver version is insufficient for CUDA runtime version'.If i used twnsorflow-gpu==1.13.1 is ok. But if i used pytorch2keras is ok on tensorflow 1.14.0.why?what is different?

gmalivenko commented 4 years ago

Hello @lawo123 .

  1. v0.0.5 is very old, please, try the recent versions
  2. 'CUDA driver version is insufficient for CUDA runtime version' means you try to use CUDA version that is differ from you driver's version. You can update GPU drivers (or restart you computer if you already have done that to apply changes)
lawo123 commented 4 years ago

Hello @lawo123 .

  1. v0.0.5 is very old, please, try the recent versions
  2. 'CUDA driver version is insufficient for CUDA runtime version' means you try to use CUDA version that is differ from you driver's version. You can update GPU drivers (or restart you computer if you already have done that to apply changes)

I used pytorch2keras is ok.but if i pip install onnx2keras it occurs error.I don't know why. I know pip will install tensorflow2.0,i uninstall it and pip install tensorflow-gpu==1.14.0. I ask that Because I want to modify some codes.thank you very much for your reply