腾讯云
操作计划
1-0ftp上传测试集、训练集zip；
如果要压缩的是个文件夹，则要加上-r参数，表示调用递归压缩
zip -r temp.zip temp
unzip all.zip
2-测试python环境
3-编译环境
yum install make gcc gcc-c++ openssl-devel zlib-devel -y;
yum -y install epel-release;
4-安装py3
wget https://www.python.org/ftp/python/3.6.5/Python-3.6.5.tgz
tar -zxvf Python-3.6.5.tgz
进入解压缩后的目录，安装配置：
cd
耗时
./configure;make;make install;
[root@VM_33_32_centos bin]# pwd
/usr/local/bin
[root@VM_33_32_centos bin]# ll -as
total 24896
4 drwxr-xr-x. 2 root root 4096 May 4 21:29 .
4 drwxr-xr-x. 13 root root 4096 May 4 21:22 ..
0 lrwxrwxrwx 1 root root 8 May 4 21:29 2to3 -> 2to3-3.6
4 -rwxr-xr-x 1 root root 101 May 4 21:29 2to3-3.6
4 -rwxr-xr-x 1 root root 242 May 4 21:28 easy_install-3.6
0 lrwxrwxrwx 1 root root 7 May 4 21:29 idle3 -> idle3.6
4 -rwxr-xr-x 1 root root 99 May 4 21:29 idle3.6
4 -rwxr-xr-x 1 root root 214 May 4 21:28 pip3
4 -rwxr-xr-x 1 root root 214 May 4 21:28 pip3.6
0 lrwxrwxrwx 1 root root 8 May 4 21:29 pydoc3 -> pydoc3.6
4 -rwxr-xr-x 1 root root 84 May 4 21:29 pydoc3.6
0 lrwxrwxrwx 1 root root 9 May 4 21:29 python3 -> python3.6
12428 -rwxr-xr-x 2 root root 12705560 May 4 21:28 python3.6
0 lrwxrwxrwx 1 root root 17 May 4 21:29 python3.6-config -> python3.6m-config
12428 -rwxr-xr-x 2 root root 12705560 May 4 21:28 python3.6m
4 -rwxr-xr-x 1 root root 3117 May 4 21:29 python3.6m-config
0 lrwxrwxrwx 1 root root 16 May 4 21:29 python3-config -> python3.6-config
0 lrwxrwxrwx 1 root root 10 May 4 21:29 pyvenv -> pyvenv-3.6
4 -rwxr-xr-x 1 root root 441 May 4 21:29 pyvenv-3.6
[root@VM_33_32_centos bin]#
./pip3 install numpy
./pip3 install pandas
./pip3 install tensorflow
验证安装成功
[root@VM_33_32_centos bin]# python3
Python 3.6.5 (default, May 4 2018, 21:28:55)
[GCC 4.8.5 20150623 (Red Hat 4.8.5-16)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import pandas as pd
>>> import tensorflow as tf
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
2018-05-04 21:39:43.001855: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
>>> print(sess.run(hello))
b'Hello, TensorFlow!'
>>>