Yi Wang

Add tutorial "How to Run"

Add tutorial "How to Contribute Models"

Make sure that the command-line tool `elasticdl` can be pip installed.

Make sure that `elasticdl` can reuse models in a Git repo and pack Docker images for running.

Hi @ogoffart , I am wondering how did you get that very long line of compiler arguments for Tensorflow? Thanks! Yi

Hi @Yangff , Thank you for the hint and the awesome work at http://yangff.coding.me/tf-doc/codebrowser/codebrowser/tensorflow ! But when I tried to follow your hint, I got an error complaining that the...

> 我个人认为比较理想的方式是serving可以直接用重构后的remote session（这里是Python的接口，如果需要高性能我们可以提供C++/Go的接口），这里是模块示意图。如果用户需要HTTP接口，可以自己在上面再包一层。 好啊！

> ## Prerequisites > - [install minikube](https://kubernetes.io/docs/tasks/tools/install-minikube/) > - [install kubectl](https://kubernetes.io/docs/tasks/tools/install-kubectl/) This section should be a step, not a section. We only need the first of the two external links;...

> https_proxy=https://YOURPROXY:PORT minikube start --kubernetes-version v1.6.4 We need to explain why we need 1.6.4. here, if we do need this specific version; or, let's explain that any recent version works,...

> ``` > mkdir > ``` > - mac > Since Minikube mounts `$Home` path by default, we recommend creating the path under `$Home` which offers the flexibility of switching...