Could not load dynamic library 'libcudart.so.10.0'; dlerror: libcudart.so.10.0: cannot open shared object file: No such file or directory

Essentially what's happening is that you do not have the appropriate CUDA Toolkit and CUDNN installed. Your drivers may be fine.

This particular error shows we are missing 10.0. However some cases can show that you are missing other versions like 10.1 or 10.2.

Before continuing I want to first say that at the time of writing this article the tensorflow plugin does not offer ARM support, that includes devices like Jetson Nano and Raspberry Pi.

You can learn to install different versions of CUDA Toolkit (and CUDNN) here.

A more descript version of this error may look like this.

2|shinobi- | cpu backend was already registered. Reusing existing backend factory.
2|shinobi- | Platform node has already been set. Overwriting the platform with [object Object].
2|shinobi- | 2020-11-17 15:57:06.939653: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: FMA
2|shinobi- | 2020-11-17 15:57:06.983770: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 4018500000 Hz
2|shinobi- | 2020-11-17 15:57:06.984985: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x52ea120 initialized for platform Host (this does not guarantee that XLA will be used). Devices:
2|shinobi- | 2020-11-17 15:57:06.985030: I tensorflow/compiler/xla/service/service.cc:176]   StreamExecutor device (0): Host, Default Version
2|shinobi- | 2020-11-17 15:57:06.987395: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcuda.so.1
2|shinobi- | 2020-11-17 15:57:07.011082: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties:
2|shinobi- | name: P106-100 major: 6 minor: 1 memoryClockRate(GHz): 1.7085
2|shinobi- | pciBusID: 0000:01:00.0
2|shinobi- | 2020-11-17 15:57:07.011224: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcudart.so.10.0'; dlerror: libcudart.so.10.0: cannot open shared object file: No such file or directory
2|shinobi- | 2020-11-17 15:57:07.011303: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcublas.so.10.0'; dlerror: libcublas.so.10.0: cannot open shared object file: No such file or directory
2|shinobi- | 2020-11-17 15:57:07.011393: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcufft.so.10.0'; dlerror: libcufft.so.10.0: cannot open shared object file: No such file or directory
2|shinobi- | 2020-11-17 15:57:07.011462: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcurand.so.10.0'; dlerror: libcurand.so.10.0: cannot open shared object file: No such file or directory
2|shinobi- | 2020-11-17 15:57:07.011529: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcusolver.so.10.0'; dlerror: libcusolver.so.10.0: cannot open shared object file: No such file or directory
2|shinobi- | 2020-11-17 15:57:07.011601: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libcusparse.so.10.0'; dlerror: libcusparse.so.10.0: cannot open shared object file: No such file or directory
2|shinobi- | 2020-11-17 15:57:07.019765: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudnn.so.7
2|shinobi- | 2020-11-17 15:57:07.019795: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1641] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
2|shinobi- | Skipping registering GPU devices...
2|shinobi- | 2020-11-17 15:57:07.111114: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1159] Device interconnect StreamExecutor with strength 1 edge matrix:
2|shinobi- | 2020-11-17 15:57:07.111152: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1165]      0
2|shinobi- | 2020-11-17 15:57:07.111161: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1178] 0:   N

Come join the Shinobi Community if you have questions! https://shinobi.community