Tensorflow 1.15 wont read CUDA on Ubuntu
By Emma Johnson •
I did: sudo apt install nvidia-cuda-toolkitand the system installed CUDA 9.1.
Important context: running nvidia-smi gives the following output:
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.Running the test package in tf gives the following outputs:
>>> tf.test.is_built_with_cuda()
True
>>> tf.test.is_built_with_gpu_support()
True
>>> tf.test.is_gpu_available(cuda_only=False,min_cuda_compute_capability=None)
2020-06-04 13:55:32.347286: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2020-06-04 13:55:32.666622: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2899885000 Hz
2020-06-04 13:55:32.669156: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x5558285d1550 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2020-06-04 13:55:32.669288: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version
2020-06-04 13:55:32.782122: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2020-06-04 13:55:33.318362: E tensorflow/stream_executor/cuda/cuda_driver.cc:318] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected
2020-06-04 13:55:33.318551: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (lenovo-ritik): /proc/driver/nvidia/version does not exist
FalseSo how do i get around this? How can I get CUDA to run and perhaps also have the system read my GPU?
Reset to default
