Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory

Dear Team,
This is my first time building a website, I was trying to use load_models from keras.models to import a CNN model in my flask backend, so I included tensorflow==2.11.0 and keras==2.11.0 in my requirements.txt file. But I keep getting this weird error that involves CUDA versions and GPU support in tensorflow, yet I don’t need GPU support for load_models. Is there something wrong with my tensorflow and keras versions or is there something I did not include in requirements.txt?
Thanks a lot!

Jan 17 11:04:39 PM 2023-01-18 04:04:39.858970: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
Jan 17 11:04:39 PM To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Jan 17 11:04:41 PM 2023-01-18 04:04:41.556740: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library ‘libcudart.so.11.0’; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory
Jan 17 11:04:41 PM 2023-01-18 04:04:41.556820: I tensorflow/compiler/xla/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Jan 17 11:04:55 PM 2023-01-18 04:04:55.273378: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library ‘libnvinfer.so.7’; dlerror: libnvinfer.so.7: cannot open shared object file: No such file or directory
Jan 17 11:04:55 PM 2023-01-18 04:04:55.352426: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library ‘libnvinfer_plugin.so.7’; dlerror: libnvinfer_plugin.so.7: cannot open shared object file: No such file or directory
Jan 17 11:04:55 PM 2023-01-18 04:04:55.352514: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.
Jan 17 11:05:21 PM 2023-01-18 04:05:21.271489: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library ‘libcuda.so.1’; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory
Jan 17 11:05:21 PM 2023-01-18 04:05:21.271597: W tensorflow/compiler/xla/stream_executor/cuda/cuda_driver.cc:265] failed call to cuInit: UNKNOWN ERROR (303)
Jan 17 11:05:21 PM 2023-01-18 04:05:21.271641: I tensorflow/compiler/xla/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (srv-cf3lg194rebfa0pg79d0-kn-00004-deployment-59d945769-x5m6c): /proc/driver/nvidia/version does not exist
Jan 17 11:05:21 PM 2023-01-18 04:05:21.272126: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
Jan 17 11:05:21 PM To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Jan 17 11:05:29 PM [2023-01-18 04:05:29 +0000] [54] [INFO] Starting gunicorn 20.1.0
Jan 17 11:05:29 PM [2023-01-18 04:05:29 +0000] [54] [INFO] Listening at: http://0.0.0.0:10000 (54)
Jan 17 11:05:29 PM [2023-01-18 04:05:29 +0000] [54] [INFO] Using worker: sync
Jan 17 11:05:29 PM [2023-01-18 04:05:29 +0000] [85] [INFO] Booting worker with pid: 85

Hi there,

From those logs it looks like the service is still booting.

I’m not familiar with TensorFlow, but the message after the missing Cuda files is:

Ignore above cudart dlerror if you do not have a GPU set up on your machine.

Current Render instances don’t have GPUs, so it looks like you can ignore the messages/warnings.

Alan

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.