外部系统调用AWS ML服务的接口 － Python

外部系统调用AWS ML服务的接口 － Python

Amazon Machine Learning


https://aws.amazon.com/machine-learning/



Steps

Step 1: Prepare Your Data

Step 2: Create a Training Datasource

Step 3: Create an ML Model

Step 4: Review the ML Model's Predictive Performance

Step 5: Use the ML Model to Generate Predictions

Step 6: Clean Up

准备数据(清洗，转换...)→选模型→检查结果→预测新数据→清理

1. 登录 EC2

2. 下载

cd  /home/ML

wget  https://github.com/awslabs/machine-learning-samples/archive/master.zip

unzip  master.zip

3. 进入

 cd  machine-learning-samples-master

[root@ip-172-30-0-89 machine-learning-samples-master]# ls k-fold-cross-validation  ml-tools-python  mobile-ios  social-media             targeted-marketing-python LICENSE                  mobile-android   README.md   targeted-marketing-java  targeted-marketing-scala

4. 测试 K-fold Cross-validation

[root@ip-172-30-0-89 k-fold-cross-validation]# ls banking.csv.schema  build_folds.py  collect_perf.py  config.py  fold.py  README.md  recipe.json  setup.sh

5. 安装 python 依赖

版本 Python 2.7.10

在 apt-get 环境下

sudo apt-get update sudo apt-get install python-pip python-virtualenv

在yum 环境下

yum install -y python27-pip  python27-virtualenv

安装的版本

[root@ip-172-30-0-89 k-fold-cross-validation]# /usr/bin/virtualenv  --version

12.0.7

[root@ip-172-30-0-89 k-fold-cross-validation]# pip --version

pip 6.1.1 from /usr/lib/python2.7/dist-packages (python 2.7)

6. 执行setup.sh 脚本，安装 boto

boto的主页在  http://boto.readthedocs.io/en/latest/getting_started.html#getting-started-with-boto

[root@ip-172-30-0-89 k-fold-cross-validation]# source  setup.sh  Not deleting local-python/bin New python executable in local-python/bin/python2.7 Also creating executable in local-python/bin/python Installing setuptools, pip...done. You are using pip version 6.0.8, however version 9.0.1 is available. You should consider upgrading via the 'pip install --upgrade pip' command. Collecting pip from https://pypi.python.org/packages/b6/ac/7015eb97dc749283ffdec1c3a88ddb8ae03b8fad0f0e611408f196358da3/pip-9.0.1-py2.py3-none-any.whl#md5=297dbd16ef53bcef0447d245815f5144   Downloading pip-9.0.1-py2.py3-none-any.whl (1.3MB)     100% |################################| 1.3MB 380kB/s  Installing collected packages: pip   Found existing installation: pip 6.0.8     Uninstalling pip-6.0.8:       Successfully uninstalled pip-6.0.8 Successfully installed pip-9.0.1 Collecting boto   Downloading boto-2.45.0-py2.py3-none-any.whl (1.3MB)     100% |████████████████████████████████| 1.4MB 877kB/s  Installing collected packages: boto Successfully installed boto-2.45.0

说明： 退出， 使用 deactivate 命令

To exit the virtual environment, type deactivate.
To clean up the dependent libraries after exiting, remove the ./local-python directory.

7. 缺少 Credentials

（后续补充）