win10+anaconda+RTX3070+tensorflow-gpu深度学习环境搭建

1、安装版本

Python 3.8.5

tf-nightly-gpu 2.5.0

Anaconda3-2020.11-Windows-x86_64

2、下载 Anaconda

Index of /anaconda/archive/ | 清华大学开源软件镜像站 | Tsinghua Open Source Mirror

3、 Anaconda Prompt中输入安装指令

conda create -n tensorflow-gpu python=3.8.5

activate tensorflow-gpu

conda install -c conda-forge cudnn

pip install tf-nightly-gpu 

国内镜像安装：

pip install tf-nightly-gpu -i https://pypi.tuna.tsinghua.edu.cn/simple
或者
pip install tensorflow-gpu==2.3.0 --default-timeout=10000 -i https://pypi.doubanio.com/simple

conda install scipy pillow

conda install jupyterlab

conda install scikit-learn

conda install matplotlib (pip install matplotlib)

conda install pandas（conda install openpyxl）

conda install -c conda-forge opencv

conda install -c conda-forge lightgbm

pip install xgboost -i https://pypi.tuna.tsinghua.edu.cn/simple/

pip install SimpleITK -i https://pypi.tuna.tsinghua.edu.cn/simple/

pip install pyradiomics -i https://pypi.tuna.tsinghua.edu.cn/simple/

jupyter notebook

 注意：conda安装太慢时，Windows 用户无法直接创建名为 .condarc 的文件，可先执行 conda config --set show_channel_urls yes 生成该文件之后再修改。

channels:
  - defaults
show_channel_urls: true
default_channels:
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/r
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/msys2
custom_channels:
  conda-forge: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  msys2: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  bioconda: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  menpo: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  pytorch: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  pytorch-lts: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud
  simpleitk: https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud

 运行 conda clean -i 清除索引缓存，保证用的是镜像站提供的索引。

4、检测是否安装成功

import tensorflow as tf
print(tf.__version__)
tf.config.list_physical_devices()

安装成功！