关于tensorflow2.5.0+cuda11.1+cudnn+python3.8以及对应torch的win64安装解决问题。

博主在论文需求下,逐步解决tensorflow GPU版与cuda、cudnn、pytorch的安装问题。通过确定显卡兼容性,安装cuda11.1,拷贝cudnn文件到指定目录,安装torch1.9.1和tensorflow2.5.0,解决缺失dll文件问题,最终实现GPU支持。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >

tf.test.is_built_with_cuda():True
tf.test.is_gpu_available():False
这个问题困扰了本人很久,因为论文需要,找了许多的解决办法,终于将出现的问题一个一个的解决了!
本来因为模型需要,我只需要运行神经网络的部分,一开始cpu计算能力完全是够的,最多就等久一点。但随着模型的深入,cpu开始崩了,于是想到了使用gpu。
最开始用的tensorflow是2.7.0的版本,当时虽然也有许多问题,但基本也能解决。但后面想调用gpu时发现超过了可以兼容的cudn。最后通过各种尝试,终于实现了tensorflow2.X+cuda+cudnn+torch的稳定版(代码放后面了)。
库包在里面了链接:百度网盘 请输入提取码icon-default.png?t=M276https://pan.baidu.com/s/1WLzTKv8eJZ7SR6rh01OujQ
提取码:uq74
--来自百度网盘超级会员V6的分享。

3)v1.10.1 ~ v1.7.02. 安装全过程(1)选择版本(2安装结果参考文章一、版本对应下表来自 pytorch 的 github 官方文档:pytorch/vision: Datasets, Transforms and Models specific to Computer Visionpytorch 安装官网:Start Locally | PyTorchpytorch 之前版本的安装命令:Previous PyTorch Versions | PyTorchtorchtorchvision 等相关库:download.pytorch.org/whl/torch_stable.html其中,命令中 "-c pytorch" 表示官方源,自己换源可以去掉。torch 版本 torchvision 版本 torchaudio 版本 支持的 Python 版本(示例) Cuda 版本2.5.1 0.20.1 2.5.1 >=3.9, <3.133.12)[9/10/11/12] 12.4/12.1/11.82.5.0 0.20.0 2.5.0 >=3.9, <3.133.1212.4/12.1/11.82.4.1 0.19.1 2.4.1 >=3.8, <3.133.12)[8/9/10/11/12] 12.4/12.1/11.82.4.0 0.19.0 2.4.0 >=3.8, <3.133.1212.4/12.1/11.82.3.1 0.18.1 2.3.1 >=3.8, <3.133.128/9/10/11/12 12.1/11.82.3.0 0.18.0 2.3.0 >=3.8, <3.133.1212.1/11.82.2.2 0.17.2 2.2.2 >=3.8, <3.12 [8/9/10/11] 12.1/11.82.2.1 0.17.1 2.2.1 >=3.8, <3.12 12.1/11.82.2.0 0.17.0 2.2.0 >=3.8, <3.12 12.1/11.82.1.2 0.16.2 2.1.2 >=3.8, <3.123.108/9/10/11 12.1/11.82.1.1 0.16.1 2.1.1 >=3.8, <3.123.1012.1/11.82.1.0 0.16.0 2.1.0 >=3.8, <3.123.1012.1/11.82.0.0 0.15.0 2.0.0 >=3.8, <3.123.8)[8/9/10/11] 11.8/11.71.13.1 0.14.1 0.13.1 >=3.7.2, <=3.103.8)[7/8/9/10] 11.7/11.61.13.0 0.14.0 0.13.0 >=3.7.2, <=3.103.811.7/11.61.12.1 0.13.1 1.12.1 >=3.7, <=3.103.8)[7/8/9/10] 11.6/11.3/10.21.12.0 0.13.0 1.12.0 >=3.7, <=3.103.811.6/11.3/10.21.11.0 0.12.0 1.11.0 >=3.7, <=3.103.811.3/10.21.10.1 0.11.2 0.10.1 >=3.6, <=3.9(3.8)[6/7/8/9] 11.3/10.21.10.0 0.11.0 0.10.0 >=3.6, <=3.9(3.811.3/10.21.9.1 0.10.1 0.9.1 >=3.6, <=3.9(3.8)[6/7/8/9] 11.1/10.21.9.0 0.10.0 0.9.0 >=3.6, <=3.9(3.811.1/10.21.8.1 0.9.1 0.8.1 >=3.6, <=3.9(3.8)[6/7/8/9] 11.1/10.21.8.0 0.9.0 0.8.0 >=3.6, <=3.9(3.811.1/10.21.7.1 0.8.2 0.7.2 >=3.6(3.6) 11.0/10.2/10.11.7.0 0.8.0 0.7.0 >=3.6(3.6) 11.0/10.2/10.1二、安装命令(pip)1. 版本(1)v2.5.1 ~ v2.0.0# v2.5.1# CUDA 12.4pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 --index-url https://download.pytorch.org/whl/cu124 -i https://pypi.tuna.tsinghua.edu.cn/simple/# CPU onlypip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 --index-url https://download.pytorch.org/whl/cpu -i https://pypi.tuna.tsinghua.edu.cn/simple/(2)v1.13.1 ~ v1.11.0# v1.13.1# CUDA 11.7pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu117 -i https://pypi.tuna.tsinghua.edu.cn/simple/# CPU onlypip install torch==1.13.1+cpu torchvision==0.14.1+cpu torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cpu -i https://pypi.tuna.tsinghua.edu.cn/simple/(3)v1.10.1 ~ v1.7.0# v1.10.1# CUDA 10.2pip install torch==1.10.1+cu102 torchvision==0.11.2+cu102 torchaudio==0.10.1 -f https://download.pytorch.org/whl/cu102/torch_stable.html -i https://pypi.tuna.tsinghua.edu.cn/simple/# CPU onlypip install torch==1.10.1+cpu torchvision==0.11.2+cpu torchaudio==0.10.1 -f https://download.pytorch.org/whl/cpu/torch_stable.html -i https://pypi.tuna.tsinghua.edu.cn/simple/2. 安装全过程(1)选择版本torch 版本 torchvision 版本 torchaudio 版本 支持的 Python 版本(示例) Cuda 版本2.1.0 0.16.0 2.1.0 >=3.8, <3.123.1012.1/11.8这里选择的框架和环境如下:torch2.1.0 | torchvision0.16.0 | torchaudio2.1.0 | python3.10 | Cuda12.1,若需要将创建的虚拟环境添加到 Jupyter Lab / Jupyter Notebook 中使用,则需要第 3-6 步,否则不用。打开 WIN + R,输入 “cmd”,进入命令行窗口,其他步骤如下:# 1. Anaconda 创建虚拟环境conda create -n torch python=3.10# 2. 激活并进入虚拟环境activate torch# 3. 安装 ipykernel pip install ipykernel -i https://pypi.tuna.tsinghua.edu.cn/simple/# 4. 安装ipykernel,将虚拟环境加入 jupyter 内核中python -m ipykernel install --name torch --display-name torch# 5. 检查新虚拟环境是否成功加入内核jupyter kernelspec list# 6. 从指定文件夹里进入 jupyterjupyter lab# 7. 安装 torch 等软件包
最新发布
04-01
自编译tensorflow1.python3.5,tensorflow1.12; 2.支持cuda10.0,cudnn7.3.1,TensorRT-5.0.2.6-cuda10.0-cudnn7.3; 3.支持mkl,无MPI; 软硬件硬件环境:Ubuntu16.04,GeForce GTX 1080 配置信息: hp@dla:~/work/ts_compile/tensorflow$ ./configure WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown". You have bazel 0.19.1 installed. Please specify the location of python. [Default is /usr/bin/python]: /usr/bin/python3 Found possible Python library paths: /usr/local/lib/python3.5/dist-packages /usr/lib/python3/dist-packages Please input the desired Python library path to use. Default is [/usr/local/lib/python3.5/dist-packages] Do you wish to build TensorFlow with XLA JIT support? [Y/n]: XLA JIT support will be enabled for TensorFlow. Do you wish to build TensorFlow with OpenCL SYCL support? [y/N]: No OpenCL SYCL support will be enabled for TensorFlow. Do you wish to build TensorFlow with ROCm support? [y/N]: No ROCm support will be enabled for TensorFlow. Do you wish to build TensorFlow with CUDA support? [y/N]: y CUDA support will be enabled for TensorFlow. Please specify the CUDA SDK version you want to use. [Leave empty to default to CUDA 10.0]: Please specify the location where CUDA 10.0 toolkit is installed. Refer to README.md for more details. [Default is /usr/local/cuda]: /usr/local/cuda-10.0 Please specify the cuDNN version you want to use. [Leave empty to default to cuDNN 7]: 7.3.1 Please specify the location where cuDNN 7 library is installed. Refer to README.md for more details. [Default is /usr/local/cuda-10.0]: Do you wish to build TensorFlow with TensorRT support? [y/N]: y TensorRT support will be enabled for TensorFlow. Please specify the location where TensorRT is installed. [Default is /usr/lib/x86_64-linux-gnu]:/home/hp/bin/TensorRT-5.0.2.6-cuda10.0-cudnn7.3/targets/x86_64-linux-gnu Please specify the locally installed NCCL version you want to use. [Default is to use https://github.com/nvidia/nccl]: Please specify a list of comma-separated Cuda compute capabilities you want to build with. You can find the compute capability of your device at: https://developer.nvidia.com/cuda-gpus. Please note that each additional compute capability significantly increases your build time and binary size. [Default is: 6.1,6.1,6.1]: Do you want to use clang as CUDA compiler? [y/N]: nvcc will be used as CUDA compiler. Please specify which gcc should be used by nvcc as the host compiler. [Default is /usr/bin/gcc]: Do you wish to build TensorFlow with MPI support? [y/N]: No MPI support will be enabled for TensorFlow. Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -march=native -Wno-sign-compare]: Would you like to interactively configure ./WORKSPACE for Android builds? [y/N]: Not configuring the WORKSPACE for Android builds. Preconfigured Bazel build configs. You can use any of the below by adding "--config=" to your build command. See .bazelrc for more details. --config=mkl # Build with MKL support. --config=monolithic # Config for mostly static monolithic build. --config=gdr # Build with GDR support. --config=verbs # Build with libverbs support. --config=ngraph # Build with Intel nGraph support. --config=dynamic_kernels # (Experimental) Build kernels into separate shared objects. Preconfigured Bazel build configs to DISABLE default on features: --config=noaws # Disable AWS S3 filesystem support. --config=nogcp # Disable GCP support. --config=nohdfs # Disable HDFS support. --config=noignite # Disable Apacha Ignite support. --config=nokafka # Disable Apache Kafka support. --config=nonccl # Disable NVIDIA NCCL support. Configuration finished 编译: hp@dla:~/work/ts_compile/tensorflow$ bazel build --config=opt --config=mkl --verbose_failures //tensorflow/tools/pip_package:build_pip_package 卸载已有tensorflow: hp@dla:~/temp$ sudo pip3 uninstall tensorflow 安装自己编译的成果: hp@dla:~/temp$ sudo pip3 install tensorflow-1.12.0-cp35-cp35m-linux_x86_64.whl
评论 1
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值