I installed CUDA 11 in my ubuntu 18.04 PC. I am currently using Tensorflow 2.4.1 and so I followed the instructions in 使用 pip 安装 TensorFlow. However I encounter the following error:-
2021-12-06 18:16:01.427080: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2.4.1
2021-12-06 18:16:03.922384: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
2021-12-06 18:16:03.923896: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1
2021-12-06 18:16:03.972384: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-12-06 18:16:03.972674: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:03:00.0 name: NVIDIA GeForce GTX 1650 computeCapability: 7.5
coreClock: 1.59GHz coreCount: 14 deviceMemorySize: 3.82GiB deviceMemoryBandwidth: 178.84GiB/s
2021-12-06 18:16:03.972693: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2021-12-06 18:16:03.990545: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11
2021-12-06 18:16:03.990614: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.11
2021-12-06 18:16:04.041156: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
2021-12-06 18:16:04.048538: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
2021-12-06 18:16:04.127700: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10
2021-12-06 18:16:04.127970: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library ‘libcusparse.so.11’; dlerror: libcusparse.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /opt/ros/melodic/lib
2021-12-06 18:16:04.129752: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8
2021-12-06 18:16:04.129778: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1757] 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 使用 pip 安装 TensorFlow for how to download and setup the required libraries for your platform.
Skipping registering GPU devices…
Num GPUs Available: 0
Tensorflow Version : 2.4.1
Ubuntu version : 18.04
CUDA version : 11.0
cudnn = 8
I dont know why “successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero” or “Could not load dynamic library ‘libcusparse.so.11’; dlerror: libcusparse.so.11” arises. I have looked through many solutions but none of them worked. Looking for some hints or solutions regarding the issue.
Thanks in advance