Linux环境下 GPU版本 jax jaxlib flax安装

葛蓓莉亚

已于 2025-01-11 22:04:25 修改

阅读量1.3k

点赞数 15

CC 4.0 BY-SA版权

文章标签： linux python

于 2025-01-11 21:18:15 首次发布

本文链接：https://blog.csdn.net/qq_42552025/article/details/145080591

1. CUDA安装判断

CUDA版本查看: cuda 符号链接指向 cuda-11.3，确认版本号为11.3

1.
nvcc --version
# nvcc: NVIDIA (R) Cuda compiler driver
# Copyright (c) 2005-2021 NVIDIA Corporation
# Built on Sun_Mar_21_19:15:46_PDT_2021
# Cuda compilation tools, release 11.3, V11.3.58
# Build cuda_11.3.r11.3/compiler.29745058_0

2.
ls -l /usr/local | grep cuda
# cuda -> cuda-11.3
# cuda-11.3

2. CUDnn安装

2.1 检查CUDnn是否安装

正常输出则已安装：确认CUDnn版本为8.6.0

1.检查 cuDNN 头文件
ls /usr/local/cuda/include | grep cudnn.h
# cudnn.h

2. 检查 cuDNN 库文件
ls /usr/local/cuda/lib64 | grep libcudnn

libcudnn_adv_infer.so
libcudnn_adv_infer.so.8
libcudnn_adv_infer.so.8.6.0
libcudnn_adv_infer_static.a
libcudnn_adv_infer_static_v8.a
libcudnn_adv_train.so
libcudnn_adv_train.so.8
libcudnn_adv_train.so.8.6.0
libcudnn_adv_train_static.a
libcudnn_adv_train_static_v8.a
libcudnn_cnn_infer.so
libcudnn_cnn_infer.so.8
libcudnn_cnn_infer.so.8.6.0
libcudnn_cnn_infer_static.a
libcudnn_cnn_infer_static_v8.a
libcudnn_cnn_train.so
libcudnn_cnn_train.so.8
libcudnn_cnn_train.so.8.6.0
libcudnn_cnn_train_static.a
libcudnn_cnn_train_static_v8.a
libcudnn_ops_infer.so
libcudnn_ops_infer.so.8
libcudnn_ops_infer.so.8.6.0
libcudnn_ops_infer_static.a
libcudnn_ops_infer_static_v8.a
libcudnn_ops_train.so
libcudnn_ops_train.so.8
libcudnn_ops_train.so.8.6.0
libcudnn_ops_train_static.a
libcudnn_ops_train_static_v8.a
libcudnn.so
libcudnn.so.8
libcudnn.so.8.6.0