tensorflow篇---tensorflow-gpu 用终端能用但是pycharm直接运行不行之解决办法

终端运行结果

2022-01-13 11:28:50.703903: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2022-01-13 11:28:52.229102: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
2022-01-13 11:28:52.233835: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1
2022-01-13 11:28:52.262129: 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
2022-01-13 11:28:52.262692: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.695GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s
2022-01-13 11:28:52.262719: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2022-01-13 11:28:52.296634: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11
2022-01-13 11:28:52.296732: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.11
2022-01-13 11:28:52.309037: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
2022-01-13 11:28:52.315870: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
2022-01-13 11:28:52.337727: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10
2022-01-13 11:28:52.346271: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11
2022-01-13 11:28:52.347597: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8
2022-01-13 11:28:52.347742: 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
2022-01-13 11:28:52.348327: 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
2022-01-13 11:28:52.351785: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0
2022-01-13 11:28:52.366287: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set
2022-01-13 11:28:52.366502: 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
2022-01-13 11:28:52.367090: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.695GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s
2022-01-13 11:28:52.367118: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2022-01-13 11:28:52.367183: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.11
2022-01-13 11:28:52.367199: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.11
2022-01-13 11:28:52.367214: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
2022-01-13 11:28:52.367229: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
2022-01-13 11:28:52.367244: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10
2022-01-13 11:28:52.367259: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.11
2022-01-13 11:28:52.367274: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.8
2022-01-13 11:28:52.367357: 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
2022-01-13 11:28:52.367979: 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
2022-01-13 11:28:52.368478: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0
2022-01-13 11:28:52.368510: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
2022-01-13 11:28:52.991253: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:
2022-01-13 11:28:52.991294: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267]      0
2022-01-13 11:28:52.991302: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0:   N
2022-01-13 11:28:52.991541: 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
2022-01-13 11:28:52.992182: 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
2022-01-13 11:28:52.992730: 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
2022-01-13 11:28:52.993255: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 15023 MB memory) -> physical GPU (device: 0, name: GeForce RTX 3090, pci bus id: 0000:01:00.0, compute capability: 8.6)

控制台直接运行结果如下

/home/yyq/.conda/envs/tf_37/bin/python /app/yyq/slifeProject/4fourth_project/tf/train/faster-rcnn-tf2-main/handlers/train.py
2022-01-13 11:29:23.797301: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
2022-01-13 11:29:23.797319: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
2022-01-13 11:29:25.139055: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
2022-01-13 11:29:25.140200: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1
2022-01-13 11:29:25.165870: 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
2022-01-13 11:29:25.166266: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce RTX 3090 computeCapability: 8.6
coreClock: 1.695GHz coreCount: 82 deviceMemorySize: 23.70GiB deviceMemoryBandwidth: 871.81GiB/s
2022-01-13 11:29:25.166362: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/yyq/.conda/envs/tf_37/lib/python3.7/site-packages/cv2/../../lib64:
2022-01-13 11:29:25.166423: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcublas.so.11'; dlerror: libcublas.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/yyq/.conda/envs/tf_37/lib/python3.7/site-packages/cv2/../../lib64:
2022-01-13 11:29:25.166464: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcublasLt.so.11'; dlerror: libcublasLt.so.11: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/yyq/.conda/envs/tf_37/lib/python3.7/site-packages/cv2/../../lib64:
2022-01-13 11:29:25.166506: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcufft.so.10'; dlerror: libcufft.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/yyq/.conda/envs/tf_37/lib/python3.7/site-packages/cv2/../../lib64:
2022-01-13 11:29:25.166544: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcurand.so.10'; dlerror: libcurand.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/yyq/.conda/envs/tf_37/lib/python3.7/site-packages/cv2/../../lib64:
2022-01-13 11:29:25.166583: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcusolver.so.10'; dlerror: libcusolver.so.10: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/yyq/.conda/envs/tf_37/lib/python3.7/site-packages/cv2/../../lib64:
2022-01-13 11:29:25.166627: 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: /home/yyq/.conda/envs/tf_37/lib/python3.7/site-packages/cv2/../../lib64:
2022-01-13 11:29:25.166663: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'libcudnn.so.8'; dlerror: libcudnn.so.8: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/yyq/.conda/envs/tf_37/lib/python3.7/site-packages/cv2/../../lib64:
2022-01-13 11:29:25.166670: 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 https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2022-01-13 11:29:25.177642: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set
2022-01-13 11:29:25.177668: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:
2022-01-13 11:29:25.177672: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267]

解决办法

1.修改~/.zshrc

将pycahrm的路径加到环境中，示例如下：

alias pycharm="bash /opt/pycharm-anaconda-2020.1.5/bin/pycharm.sh"

刷新生效：

source ~/.zshrc

2.修改pycharm中的环境变量

选择pycharm 菜单栏Run ——> Run-Edit Configurations ——> Environment variables ——> 将cuda的路径加进去 例如：LD_LIBRARY_PATH=/usr/local/cuda-11.2/targets/x86_64-linux/lib

这样就解决啦！！！！