cpu、gpu 安装框架pytorch，cntk,theano及测试

一，cpu 下安装

tensorflow
conda env list
source activate tensorflow
直接安装相应版本

python
import tensorflow as tf
tf.version 1.11.0

keras 直接安装

conda env list
source activate keras
import keras 2.2.2
print(keras.version)
import tensorflow as tf
tf.version

pytorch

import torch
print(torch.version)
print(torch.cuda.device_count())
print(torch.cuda.is_available())

cntk
/root/anaconda3/bin/conda env list
source activate cntk-py35

python 3.5.6
export PATH=/root/anaconda3/bin:$PATH
python -c "import cntk; print(cntk.version)"

theano

caffe2
python 3.6.9
import caffe2

安装
conda create -n caffe2 python=3.6
conda activate caffe2
conda install pytorch-nightly-cpu -c pytorch -n caffe2

python -c 'from caffe2.python import core' 2>/dev/null && echo "Success" || echo "Failure"
报错:
pip install protobuf
pip install future

参考官网安装即可

gpu

tensorflow-gpu:1.11.0 python 3.5

export PATH=/root/anaconda3/bin:$PATH
source activate tensorflow

keras
export PATH=/root/anaconda3/bin:$PATH
conda env list
source activate keras
python3.5

nvidia-docker run -it --rm pytorch-gpu:1.1.0 /bin/bash
pytorch
[root@191ddd30d4ae /]# python
Python 3.6.9 |Anaconda, Inc.| (default, Jul 30 2019, 19:07:31)
[GCC 7.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.

import torch
print(torch.version)
1.1.0

print(torch.cuda.device_count())
1

print(torch.cuda.is_available())
True

cntk

source activate cntk-py35 python3.5

python -c "import cntk; print(cntk.version)"
2.4

theano

gpu-theano-in-use:1.0.4 python2.7

source activate theano
python test.py

import theano
/root/anaconda3/envs/theano/lib/python2.7/site-packages/theano/gpuarray/dnn.py:184: UserWarning: Your cuDNN version is more recent than Theano. If you encounter problems, try updating Theano or downgrading cuDNN to a version >= v5 and <= v7.
warnings.warn("Your cuDNN version is more recent than "
Using cuDNN version 7603 on context None
Mapped name None to device cuda: GeForce GTX 960M (0000:01:00.0)

theano.version
u'1.0.4'

https://www.jianshu.com/p/4cc75a79dce9
Linux下安装miniconda
在官网下载miniconda3
执行：bash Miniconda3-latest-Linux-x86_64.sh 　
－vim ~/.bashrc
－export PATH=~/anaconda3/bin:$PATH
－source ~/.bashrc
创建虚拟环境并安装theano
基于python2.7创建一个名为theano的环境
conda create --name theano python=2.7
进入虚拟环境： source activate theano
－使用conda安装：conda install numpy scipy mkl
pip install parameterized
conda install theano pygpu

       －使用pip安装：pip install Theano

测试参考官网文档

caffe2
看官网文档安装
https://caffe2.ai/docs/getting-started.html?platform=ubuntu&configuration=compile

https://blog.csdn.net/qq_35451572/article/details/79428167

cmake
-DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-9.0
-DCUDNN_ROOT_DIR=/usr/local/cuda

To check if Caffe2 build was successful

python -c 'from caffe2.python import core' 2>/dev/null && echo "Success" || echo "Failure"

To check if Caffe2 GPU build was successful

This must print a number > 0 in order to use Detectron

python -c 'from caffe2.python import workspace; print(workspace.NumCudaDevices())'

参考
https://blog.csdn.net/Yan_Joy/article/details/70241319

https://www.nvidia.com/en-gb/data-center/gpu-accelerated-applications/caffe2/
https://blog.csdn.net/qq_35451572/article/details/79428167
https://blog.csdn.net/qq_16525279/article/details/79724728
https://blog.csdn.net/y_f_raquelle/article/details/83278953
https://www.cnblogs.com/nanzhao/p/9596844.html

附:conda常用

1. conda env list 或 conda info -e 查看当前存在哪些虚拟环境

2. conda update conda 检查更新当前conda

3. conda update --all 更新本地已安装的包

4. conda create -n your_env_name python=X.X（2.7、3.6等） anaconda 命令创建python版本为X.X、名字为your_env_name的虚拟环境。your_env_name文件可以在Anaconda安装目录envs文件下找到。

5. Windows: activate your_env_name(虚拟环境名称) 激活虚拟环境

6. conda install -n your_env_name [package] 安装package到your_env_name中

7. linux: source deactivate Windows: deactivate 关闭虚拟环境

8. conda remove -n your_env_name(虚拟环境名称) --all 删除虚拟环境

9. conda remove --name your_env_name package_name 删除环境中的某个