剑客
关注科技互联网

Moving machine learning from practice to production

作者:Ramanan Balakrishnan

原文: https://engineering.semantics3.com/2016/11/13/machine-learning-practice-to-production/

我爱机器学习(52ml.net)编者按:机器学习产品化时需要注意哪些问题?数据获取、数据预处理、编程语言/框架选择、训练模型、离线或实时、内嵌或接口、RPC或其它、预测监控、log处理、Online Learning等。

Garbage in, garbage out

Do I have a reliable source of data? Where do I obtain my dataset?

Transforming data to input

What pre-processing steps are required? How do I normalize my data before using with my algorithms?

Now, let’s begin?

Which language/framework do I use? Python, R, Java, C++? Caffe, Torch, Theano, Tensorflow, DL4J?

Training models

How do I train my models? Should I buy GPUs, custom hardware, or ec2 (spot?) instances? Can I parallelize them for speed?

No system is an island

Do I need to make batched or real-time predictions? Embedded models or interfaces? RPC or REST?

Monitoring performance

How do I keep track of my predictions? Do I log my results to a database? What about online learning?

Moving machine learning from practice to production

欢迎加入我爱机器学习QQ6群: 337537549

Moving machine learning from practice to production

微信扫一扫,关注我爱机器学习公众号

微博:我爱机器学习

分享到:更多 ()

评论 抢沙发

  • 昵称 (必填)
  • 邮箱 (必填)
  • 网址