Yi Wang

https://github.com/sql-machine-learning/sqlflow/blob/140ded7edce80cddad58d3171d4f9dbe16b139be/pkg/server/sqlflowserver.go#L63 and https://github.com/sql-machine-learning/sqlflow/blob/140ded7edce80cddad58d3171d4f9dbe16b139be/pkg/server/sqlflowserver.go#L52 Should we keep the fetching log feature in either one of the above functions?

Add section Prediction in high-level-design.md

3

following the discussion here https://github.com/wangkuiyi/elasticdl/pull/815#discussion_r298789944

Improve Docker build speed in CI

1

I see that in the .travis.yaml file of SQLFlow, there is a line https://github.com/sql-machine-learning/sqlflow/blob/a033b42926e4d0ccf63c044e414aa99eb9dfcfaf/.travis.yml#L38 It seems that our .travis.yaml doesn't have such a line https://github.com/sql-machine-learning/elasticdl/blob/3a157c1c2105a2802b160d030679a3d4c66738ea/.travis.yml#L25-L31 If we add such a...

Currently, the master is in Python, because the master also works as the parameter server that calls TensorFlow's Python API to aggregate gradients and updates the global model. In our...

Complete the minikube tutorial

13

The [Running PaddlePaddle Cloud on Minikube tutorial](https://github.com/PaddlePaddle/cloud/blob/develop/doc/run_on_minikube.md) doesn't explain how to verify the successful configuration of the minikube cluster. I'd thought that we could follow this tutorial when we develop...

Why Kubernetes version 1.6.4

2

In the document https://github.com/PaddlePaddle/cloud/develop/doc/run_on_minikube.md, it requires Kubernetes version 1.6.4, but doesn't explain why that specific version. Is there any reason we have to stick to 1.6.4?

Restructure the doc directory

1

`doc/` include the following 1. `design` 1. `tutorials` 1. `benchmark`