Wu Yi comments

Results 110 comments of


                                            Wu Yi

helm chart

Please feel free to try and join https://github.com/alauda/kubeflow-chart

Support parallel for operations, like data parallel training, model parallel training etc

@akchinSTC Can you please checkout this feature

Support parallel for operations, like data parallel training, model parallel training etc

> > @akchinSTC Can you please checkout this feature > > What is the current status of this PR? I still see the `work in progress` tag.. This PR is...

WIP: Support Tensorflow distributed training in kfp workflow

@akchinSTC Thanks for the information. Actually, I'm looking for a more generic implementation, like workflow `ParallelFor` function, not only for distributed data parallel training, but also for parallel data processing...

go test -v panic

Are you running `go test` in the docker container `sqlflow/gohive:dev` or on somewhere else?

Document structure design

Test model -> model validation Prediction services(Offline) -> batch predict on a data table

Some problems in model development

@Echo9573 doing data pre-processing using SQL is currently under the design phase, please take a look at https://github.com/sql-machine-learning/elasticdl/pull/1477 and we'll appreciate your comments and advises.

Unify model zoo implementation of SQLFlow and ElasticDL

@terrytangyuan > Where is the naming convention for default_optimizer from? If necessary, we can change this in ElasticDL. ElasticDL also supports passing different names for optimizer/loss/etc. Either to change SQLFlow...

Unify model zoo implementation of SQLFlow and ElasticDL

ps: also need to support `eval_metrics_fn`

Unify model zoo implementation of SQLFlow and ElasticDL

By changing the default function name to define optimizer and loss from `default_optimizer` and `default_loss` to `optimizer` and `loss` will cause an error when predicting: ``` Traceback (most recent call...