Tensorflow в WSL: невозможно зарегистрировать завод Cufft - один уже зарегистрирован ⇐ Python

[Anonymous]

Я пытаюсь получить Tensorflow с Cuda. Поскольку Windows больше не поддерживается, я установил его в WSL. < /P>
Я установил все необходимые инструменты: < /p>

• nvidia-smi дает мою правильную информацию GPU как в WSL, так и в хост -системе (отчеты V 575.51.03
• nvcc в WSL говорит, что это выпуск 12.0, v12.0.140
• cuda-toolkit уже самая новая версия (12.9.0-1) в соответствии с app
• tensorflow-версия 2.19.0

Я получаю эти ошибки:
2025-05-26 21:22:32.067575: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:467] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
E0000 00:00:1748287352.084328 14362 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
E0000 00:00:1748287352.089101 14362 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
W0000 00:00:1748287352.104529 14362 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1748287352.104570 14362 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1748287352.104576 14362 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
W0000 00:00:1748287352.104579 14362 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.
2025-05-26 21:22:32.110219: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 AVX512F AVX512_VNNI AVX512_BF16 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
W0000 00:00:1748287354.131098 14362 gpu_device.cc:2341] 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...


Подробнее здесь: https://stackoverflow.com/questions/796 ... been-regis