GitLab

1.Add kubeflow in experiment config document
2.Add AKS in kubeflow document

* [Kubeflow training service] fix bug that wrongly split kube delete cmd into 2 lines

* Adjust white space

* fix bug

* add docs

* Support to show 2 logPath

* fix lint

* Update trial status color

* update doc for "write trial"

* fix link

* issue 414

* Kubeflow training service documentation, v1

* Fix typos based on comments

1.Refactor nnictl information when validateion error.
2.Set kubernetesServer as optional.

* [Trial keeper refactor] refactor trial keeper stdout output

* Add codeDir file count validation for setClusterConfig

* fix a small bug if find command is not installed

* Remove codeDir validation for local training service

* Remove useless import

* Fix trial job start time

* updates

* updates

* Fix bug

* fix lint

* [PAI training service] Support virtual cluster config

* fix a small bug to convert virtualCluster to string

Support aks of kuberflow training service
Support nnictl set nniManagerIp

* fix bugs due to ts.tailstream (#273)

* update Makefile for mac support, wait for aka.ms support

* refix Makefile for colorful echo

* update Makefile with shorturl

* fix false fail on mac webui

* fix cross os remote tmpdir issue

* add readonly to RemoteMachineTrainingService.remoteOS

* fix var name for PR 386

* Rest retrieve multiple final results for multiphase job

* updates

* add one more trial job status, EARLY_STOPPED

* fix datastore/nnimanager/mockeddatastore. test/webui/metrics_reader not done. USER_TO_CANCEL

* fix bug

* modifications based on Deshui's comments

* fix bug

* fix bug in remote mode

Add nniManager Ip in nnictl, pai TrainingService and kubeflow TrainingService.
If users set nniManagerIp, pai and kubeflow will use this ip instead of using getIPV4() function.
Web UI will also use this nniManagerIp.

* Adjust sleep position for sdk_test.py

* Exit dispather process if receive Terminate command

* Add comment for sleep change in sdk_test.py

In nnictl, classArgs is not required, now set it as optional for some kind of tuner and assessor may not require classArgs.

* Use different output folder for ps and worker

* Add cuda_visible_devices env var if gpuNum is 0

* add gpuNum check for local TS

* set CUDA_VISIBLE_DEVICES to empty string when gpuNum is 0

* remove redundency code

[Kubeflow training service] Update kubeflow exp job config schema to support distributed training (#387)

* Support distributed training on tf-operator, for worker and ps

* Update validation rule for kubeflow config

* small code refactor adjustment for private methods

* Use different output folder for ps and worker

* Asynchronous dispatcher

* updates

* updates

* updates

* updates

* fix sdk's unittest and add medianstop, batchtuner to ci

* fix sdk's unittest and add medianstop, batchtuner to ci

* remove debug info

* update azure-pipelines

* remove useless code

* add some checks

* fix pylint

* update ci test

* update ci

* Kubeflow TrainingService support, v1 (#373)

1. Create new Training Service: kubeflow trainning service, use 'kubectl' and kubeflow tfjobs CRD to submit and manage jobs
2. Update nni python SDK to support new kubeflow platform
3. Update nni python SDK's get_sequende_id() implementation, read NNI_TRIAL_SEQ_ID env variable, instead of reading .nni/sequence_id file
4. This version only supports Tensorflow operator. Will add more operators' support in future versions

* Add more tooltip in default metric graph and fix bug

* update

Fix "nnictl stop"

* add gridsearch tuner

* add gridsearchtuner

* add gridsearchtuner

* add gridsearchtuner

* update gridsearch tuner

* update gridsearch tuner

* update gridsearch tuner

* update gridsearch tuner

* update gridsearch tuner

* update gridsearch tuner

* update gridsearch tuner

* update gridsearch and pylint