Kubeflow-training

[Full Changelog](https://github.com/kubeflow/training-operator/compare/v1.3.0...HEAD)

**Merged pull requests:**

- extends path in \_\_init\_\_.py for SDK correctly [\1531](https://github.com/kubeflow/training-operator/pull/1531) ([cakeislife100](https://github.com/cakeislife100))
- Update manifests with latest image tag [\1527](https://github.com/kubeflow/training-operator/pull/1527) ([johnugeorge](https://github.com/johnugeorge))
- add option for mpi kubectl delivery [\1525](https://github.com/kubeflow/training-operator/pull/1525) ([zw0610](https://github.com/zw0610))
- restore option namespace in launch arguments [\1524](https://github.com/kubeflow/training-operator/pull/1524) ([zw0610](https://github.com/zw0610))
- remove unused scripts [\1521](https://github.com/kubeflow/training-operator/pull/1521) ([zw0610](https://github.com/zw0610))
- remove ChanYiLin from approvers [\1513](https://github.com/kubeflow/training-operator/pull/1513) ([ChanYiLin](https://github.com/ChanYiLin))
- add StacktraceLevel for zapr [\1512](https://github.com/kubeflow/training-operator/pull/1512) ([qiankunli](https://github.com/qiankunli))
- add unit tests for tensorflow controller [\1511](https://github.com/kubeflow/training-operator/pull/1511) ([zw0610](https://github.com/zw0610))
- add the example of MPIJob [\1508](https://github.com/kubeflow/training-operator/pull/1508) ([hackerboy01](https://github.com/hackerboy01))
- Added 2022 roadmap and migrated previous roadmap from kubeflow/common [\1500](https://github.com/kubeflow/training-operator/pull/1500) ([terrytangyuan](https://github.com/terrytangyuan))
- Fix a typo in mpi controller log [\1495](https://github.com/kubeflow/training-operator/pull/1495) ([LuBingtan](https://github.com/LuBingtan))
- feat\(pytorch\): Add init container config to avoid DNS lookup failure [\1493](https://github.com/kubeflow/training-operator/pull/1493) ([gaocegege](https://github.com/gaocegege))
- chore: Fix GitHub Actions script [\1491](https://github.com/kubeflow/training-operator/pull/1491) ([tenzen-y](https://github.com/tenzen-y))
- chore: Fix missspell in tfjob [\1490](https://github.com/kubeflow/training-operator/pull/1490) ([tenzen-y](https://github.com/tenzen-y))
- chore: Update OWNERS [\1489](https://github.com/kubeflow/training-operator/pull/1489) ([gaocegege](https://github.com/gaocegege))
- Bump jinja2 from 2.10.1 to 2.11.3 in /py/kubeflow/tf\_operator [\1487](https://github.com/kubeflow/training-operator/pull/1487) ([dependabot[bot]](https://github.com/apps/dependabot))
- fix comments for mpi-controller [\1485](https://github.com/kubeflow/training-operator/pull/1485) ([hackerboy01](https://github.com/hackerboy01))
- add expectation-related functions for other resources used in mpi-controller [\1484](https://github.com/kubeflow/training-operator/pull/1484) ([zw0610](https://github.com/zw0610))
- Add MPI job to README now that it's supported [\1480](https://github.com/kubeflow/training-operator/pull/1480) ([terrytangyuan](https://github.com/terrytangyuan))
- add mpi doc [\1477](https://github.com/kubeflow/training-operator/pull/1477) ([zw0610](https://github.com/zw0610))
- Set Go version of base image to 1.17 [\1476](https://github.com/kubeflow/training-operator/pull/1476) ([tenzen-y](https://github.com/tenzen-y))
- update label for tf-controller [\1474](https://github.com/kubeflow/training-operator/pull/1474) ([zw0610](https://github.com/zw0610))
- Add Akuity to the list of adopters [\1473](https://github.com/kubeflow/training-operator/pull/1473) ([terrytangyuan](https://github.com/terrytangyuan))
- Add PR template with doc checklist [\1470](https://github.com/kubeflow/training-operator/pull/1470) ([andreyvelich](https://github.com/andreyvelich))
- Add e2e failure debugging guidance [\1469](https://github.com/kubeflow/training-operator/pull/1469) ([Jeffwan](https://github.com/Jeffwan))
- chore: Add .gitattributes to ignore Jsonnet test code for linguist [\1463](https://github.com/kubeflow/training-operator/pull/1463) ([terrytangyuan](https://github.com/terrytangyuan))
- Migrate additional examples from xgboost-operator [\1461](https://github.com/kubeflow/training-operator/pull/1461) ([terrytangyuan](https://github.com/terrytangyuan))
- Minor edits to README.md [\1460](https://github.com/kubeflow/training-operator/pull/1460) ([terrytangyuan](https://github.com/terrytangyuan))
- add mpi-operator\(v1\) to the unified operator [\1457](https://github.com/kubeflow/training-operator/pull/1457) ([hackerboy01](https://github.com/hackerboy01))
- fix tfjob status when enableDynamicWorker set true [\1455](https://github.com/kubeflow/training-operator/pull/1455) ([zw0610](https://github.com/zw0610))
- feat\(pytorch\): Support elastic training [\1453](https://github.com/kubeflow/training-operator/pull/1453) ([gaocegege](https://github.com/gaocegege))
- fix: generate printer columns for job crds [\1451](https://github.com/kubeflow/training-operator/pull/1451) ([henrysecond1](https://github.com/henrysecond1))
- Fix README typo [\1450](https://github.com/kubeflow/training-operator/pull/1450) ([davidxia](https://github.com/davidxia))
- consistent naming for better readability [\1449](https://github.com/kubeflow/training-operator/pull/1449) ([pramodrj07](https://github.com/pramodrj07))
- Fix set scheduler error [\1448](https://github.com/kubeflow/training-operator/pull/1448) ([qiankunli](https://github.com/qiankunli))
- Add CI to run the tests for Go [\1440](https://github.com/kubeflow/training-operator/pull/1440) ([tenzen-y](https://github.com/tenzen-y))
- fix: Add missing retrying package that failed the import [\1439](https://github.com/kubeflow/training-operator/pull/1439) ([terrytangyuan](https://github.com/terrytangyuan))
- Generate a single `swagger.json` file for all frameworks [\1437](https://github.com/kubeflow/training-operator/pull/1437) ([alembiewski](https://github.com/alembiewski))
- Update links and files with the new URL [\1434](https://github.com/kubeflow/training-operator/pull/1434) ([andreyvelich](https://github.com/andreyvelich))
- chore: update CHANGELOG.md [\1432](https://github.com/kubeflow/training-operator/pull/1432) ([Jeffwan](https://github.com/Jeffwan))
- Add acknowledgement section in README to credit all contributors [\1422](https://github.com/kubeflow/training-operator/pull/1422) ([terrytangyuan](https://github.com/terrytangyuan))
- Add Cisco to Adopters List [\1421](https://github.com/kubeflow/training-operator/pull/1421) ([andreyvelich](https://github.com/andreyvelich))
- Add Python SDK for Kubeflow Training Operator [\1420](https://github.com/kubeflow/training-operator/pull/1420) ([alembiewski](https://github.com/alembiewski))
- docs: Move myself to approvers [\1419](https://github.com/kubeflow/training-operator/pull/1419) ([terrytangyuan](https://github.com/terrytangyuan))
- fix hyperlinks in the 'overview' section [\1418](https://github.com/kubeflow/training-operator/pull/1418) ([pramodrj07](https://github.com/pramodrj07))
- docs: Migrate adopters of all operators to this repo [\1417](https://github.com/kubeflow/training-operator/pull/1417) ([terrytangyuan](https://github.com/terrytangyuan))
- Feature/support pytorchjob set queue of volcano [\1415](https://github.com/kubeflow/training-operator/pull/1415) ([qiankunli](https://github.com/qiankunli))
- Bump controller-tools to 0.6.0 and enable GenerateEmbeddedObjectMeta [\1409](https://github.com/kubeflow/training-operator/pull/1409) ([Jeffwan](https://github.com/Jeffwan))
- Update scripts to generate sdk for all frameworks [\1389](https://github.com/kubeflow/training-operator/pull/1389) ([Jeffwan](https://github.com/Jeffwan))

**Closed issues:**

- Question: What is the recommended way for Data Scientists to run a distributed training job [\1535](https://github.com/kubeflow/training-operator/issues/1535)
- Restore KUBEFLOW\_NAMESPACE options [\1522](https://github.com/kubeflow/training-operator/issues/1522)
- Improve test coverage [\1497](https://github.com/kubeflow/training-operator/issues/1497)
- swagger.json missing Pytorchjob.Spec.ElasticPolicy [\1483](https://github.com/kubeflow/training-operator/issues/1483)
- \[bug\] Missing init container in PyTorchJob [\1482](https://github.com/kubeflow/training-operator/issues/1482)
- PytorchJob DDP training will stop if I delete a worker pod [\1478](https://github.com/kubeflow/training-operator/issues/1478)
- Write down e2e failure debug process [\1467](https://github.com/kubeflow/training-operator/issues/1467)
- How can i add the Priorityclass to the TFjob？ [\1466](https://github.com/kubeflow/training-operator/issues/1466)
- github.com/go-logr/zapr.\(\*zapLogger\).Error [\1444](https://github.com/kubeflow/training-operator/issues/1444)
- Display coverage % in GitHub actions list [\1442](https://github.com/kubeflow/training-operator/issues/1442)
- Add Go test to CI [\1436](https://github.com/kubeflow/training-operator/issues/1436)
- Podgroup is constantly created and deleted after tfjob is success or failure [\1426](https://github.com/kubeflow/training-operator/issues/1426)
- Cut official release of 1.3.0 [\1425](https://github.com/kubeflow/training-operator/issues/1425)
- Add "not maintained" notice to other operator repos [\1423](https://github.com/kubeflow/training-operator/issues/1423)
- Fail to install tf-operator in minikube because of the version of kubectl/kustomize [\1381](https://github.com/kubeflow/training-operator/issues/1381)
- Python SDK for Kubeflow Training Operator [\1380](https://github.com/kubeflow/training-operator/issues/1380)
- Rename this repo [\1348](https://github.com/kubeflow/training-operator/issues/1348)
- Universal Operator Phase III: Graduate operator to production grade [\1318](https://github.com/kubeflow/training-operator/issues/1318)

[Full Changelog](https://github.com/kubeflow/training-operator/compare/v1.3.0...v1.4.0-rc.0)

**Features and improvements:**

- Display coverage % in GitHub actions list [\1442](https://github.com/kubeflow/training-operator/issues/1442)
- Add Go test to CI [\1436](https://github.com/kubeflow/training-operator/issues/1436)

**Fixed bugs:**

- \[bug\] Missing init container in PyTorchJob [\1482](https://github.com/kubeflow/training-operator/issues/1482)
- Fail to install tf-operator in minikube because of the version of kubectl/kustomize [\1381](https://github.com/kubeflow/training-operator/issues/1381)

**Closed issues:**

- Restore KUBEFLOW\_NAMESPACE options [\1522](https://github.com/kubeflow/training-operator/issues/1522)
- Improve test coverage [\1497](https://github.com/kubeflow/training-operator/issues/1497)
- swagger.json missing Pytorchjob.Spec.ElasticPolicy [\1483](https://github.com/kubeflow/training-operator/issues/1483)
- PytorchJob DDP training will stop if I delete a worker pod [\1478](https://github.com/kubeflow/training-operator/issues/1478)
- Write down e2e failure debug process [\1467](https://github.com/kubeflow/training-operator/issues/1467)
- How can i add the Priorityclass to the TFjob？ [\1466](https://github.com/kubeflow/training-operator/issues/1466)
- github.com/go-logr/zapr.\(\*zapLogger\).Error [\1444](https://github.com/kubeflow/training-operator/issues/1444)
- Podgroup is constantly created and deleted after tfjob is success or failure [\1426](https://github.com/kubeflow/training-operator/issues/1426)
- Cut official release of 1.3.0 [\1425](https://github.com/kubeflow/training-operator/issues/1425)
- Add "not maintained" notice to other operator repos [\1423](https://github.com/kubeflow/training-operator/issues/1423)
- Python SDK for Kubeflow Training Operator [\1380](https://github.com/kubeflow/training-operator/issues/1380)

**Merged pull requests:**

- Update manifests with latest image tag [\1527](https://github.com/kubeflow/training-operator/pull/1527) ([johnugeorge](https://github.com/johnugeorge))
- add option for mpi kubectl delivery [\1525](https://github.com/kubeflow/training-operator/pull/1525) ([zw0610](https://github.com/zw0610))
- restore option namespace in launch arguments [\1524](https://github.com/kubeflow/training-operator/pull/1524) ([zw0610](https://github.com/zw0610))
- remove unused scripts [\1521](https://github.com/kubeflow/training-operator/pull/1521) ([zw0610](https://github.com/zw0610))
- remove ChanYiLin from approvers [\1513](https://github.com/kubeflow/training-operator/pull/1513) ([ChanYiLin](https://github.com/ChanYiLin))
- add StacktraceLevel for zapr [\1512](https://github.com/kubeflow/training-operator/pull/1512) ([qiankunli](https://github.com/qiankunli))
- add unit tests for tensorflow controller [\1511](https://github.com/kubeflow/training-operator/pull/1511) ([zw0610](https://github.com/zw0610))
- add the example of MPIJob [\1508](https://github.com/kubeflow/training-operator/pull/1508) ([hackerboy01](https://github.com/hackerboy01))
- Added 2022 roadmap and migrated previous roadmap from kubeflow/common [\1500](https://github.com/kubeflow/training-operator/pull/1500) ([terrytangyuan](https://github.com/terrytangyuan))
- Fix a typo in mpi controller log [\1495](https://github.com/kubeflow/training-operator/pull/1495) ([LuBingtan](https://github.com/LuBingtan))
- feat\(pytorch\): Add init container config to avoid DNS lookup failure [\1493](https://github.com/kubeflow/training-operator/pull/1493) ([gaocegege](https://github.com/gaocegege))
- chore: Fix GitHub Actions script [\1491](https://github.com/kubeflow/training-operator/pull/1491) ([tenzen-y](https://github.com/tenzen-y))
- chore: Fix missspell in tfjob [\1490](https://github.com/kubeflow/training-operator/pull/1490) ([tenzen-y](https://github.com/tenzen-y))
- chore: Update OWNERS [\1489](https://github.com/kubeflow/training-operator/pull/1489) ([gaocegege](https://github.com/gaocegege))
- Bump jinja2 from 2.10.1 to 2.11.3 in /py/kubeflow/tf\_operator [\1487](https://github.com/kubeflow/training-operator/pull/1487) ([dependabot[bot]](https://github.com/apps/dependabot))
- fix comments for mpi-controller [\1485](https://github.com/kubeflow/training-operator/pull/1485) ([hackerboy01](https://github.com/hackerboy01))
- add expectation-related functions for other resources used in mpi-controller [\1484](https://github.com/kubeflow/training-operator/pull/1484) ([zw0610](https://github.com/zw0610))
- Add MPI job to README now that it's supported [\1480](https://github.com/kubeflow/training-operator/pull/1480) ([terrytangyuan](https://github.com/terrytangyuan))
- add mpi doc [\1477](https://github.com/kubeflow/training-operator/pull/1477) ([zw0610](https://github.com/zw0610))
- Set Go version of base image to 1.17 [\1476](https://github.com/kubeflow/training-operator/pull/1476) ([tenzen-y](https://github.com/tenzen-y))
- update label for tf-controller [\1474](https://github.com/kubeflow/training-operator/pull/1474) ([zw0610](https://github.com/zw0610))
- Add Akuity to the list of adopters [\1473](https://github.com/kubeflow/training-operator/pull/1473) ([terrytangyuan](https://github.com/terrytangyuan))
- Add PR template with doc checklist [\1470](https://github.com/kubeflow/training-operator/pull/1470) ([andreyvelich](https://github.com/andreyvelich))
- Add e2e failure debugging guidance [\1469](https://github.com/kubeflow/training-operator/pull/1469) ([Jeffwan](https://github.com/Jeffwan))
- chore: Add .gitattributes to ignore Jsonnet test code for linguist [\1463](https://github.com/kubeflow/training-operator/pull/1463) ([terrytangyuan](https://github.com/terrytangyuan))
- Migrate additional examples from xgboost-operator [\1461](https://github.com/kubeflow/training-operator/pull/1461) ([terrytangyuan](https://github.com/terrytangyuan))
- Minor edits to README.md [\1460](https://github.com/kubeflow/training-operator/pull/1460) ([terrytangyuan](https://github.com/terrytangyuan))
- add mpi-operator\(v1\) to the unified operator [\1457](https://github.com/kubeflow/training-operator/pull/1457) ([hackerboy01](https://github.com/hackerboy01))
- fix tfjob status when enableDynamicWorker set true [\1455](https://github.com/kubeflow/training-operator/pull/1455) ([zw0610](https://github.com/zw0610))
- feat\(pytorch\): Support elastic training [\1453](https://github.com/kubeflow/training-operator/pull/1453) ([gaocegege](https://github.com/gaocegege))
- fix: generate printer columns for job crds [\1451](https://github.com/kubeflow/training-operator/pull/1451) ([henrysecond1](https://github.com/henrysecond1))
- Fix README typo [\1450](https://github.com/kubeflow/training-operator/pull/1450) ([davidxia](https://github.com/davidxia))
- consistent naming for better readability [\1449](https://github.com/kubeflow/training-operator/pull/1449) ([pramodrj07](https://github.com/pramodrj07))
- Fix set scheduler error [\1448](https://github.com/kubeflow/training-operator/pull/1448) ([qiankunli](https://github.com/qiankunli))
- Add CI to run the tests for Go [\1440](https://github.com/kubeflow/training-operator/pull/1440) ([tenzen-y](https://github.com/tenzen-y))
- fix: Add missing retrying package that failed the import [\1439](https://github.com/kubeflow/training-operator/pull/1439) ([terrytangyuan](https://github.com/terrytangyuan))
- Generate a single `swagger.json` file for all frameworks [\1437](https://github.com/kubeflow/training-operator/pull/1437) ([alembiewski](https://github.com/alembiewski))
- Update links and files with the new URL [\1434](https://github.com/kubeflow/training-operator/pull/1434) ([andreyvelich](https://github.com/andreyvelich))
- chore: update CHANGELOG.md [\1432](https://github.com/kubeflow/training-operator/pull/1432) ([Jeffwan](https://github.com/Jeffwan))
- Add acknowledgement section in README to credit all contributors [\1422](https://github.com/kubeflow/training-operator/pull/1422) ([terrytangyuan](https://github.com/terrytangyuan))
- Add Cisco to Adopters List [\1421](https://github.com/kubeflow/training-operator/pull/1421) ([andreyvelich](https://github.com/andreyvelich))
- Add Python SDK for Kubeflow Training Operator [\1420](https://github.com/kubeflow/training-operator/pull/1420) ([alembiewski](https://github.com/alembiewski))
- docs: Move myself to approvers [\1419](https://github.com/kubeflow/training-operator/pull/1419) ([terrytangyuan](https://github.com/terrytangyuan))
- fix hyperlinks in the 'overview' section [\1418](https://github.com/kubeflow/training-operator/pull/1418) ([pramodrj07](https://github.com/pramodrj07))
- docs: Migrate adopters of all operators to this repo [\1417](https://github.com/kubeflow/training-operator/pull/1417) ([terrytangyuan](https://github.com/terrytangyuan))
- Feature/support pytorchjob set queue of volcano [\1415](https://github.com/kubeflow/training-operator/pull/1415) ([qiankunli](https://github.com/qiankunli))
- Bump controller-tools to 0.6.0 and enable GenerateEmbeddedObjectMeta [\1409](https://github.com/kubeflow/training-operator/pull/1409) ([Jeffwan](https://github.com/Jeffwan))
- Update scripts to generate sdk for all frameworks [\1389](https://github.com/kubeflow/training-operator/pull/1389) ([Jeffwan](https://github.com/Jeffwan))

More background can be found in design doc [All-in-one Kubeflow Training Operator](https://docs.google.com/document/d/1x1JPDQfDMIbnoQRftDH1IzGU0qvHGSU4W6Jl4rJLPhI/edit)

Install Kubeflow training operator by running:


kubectl apply -k "github.com/kubeflow/tf-operator.git/manifests/overlays/standalone?ref=v1.3.0-alpha.2"

> require kubectl >= 1.21.x

[v1.3.0-rc.2](https://github.com/kubeflow/tf-operator/tree/v1.3.0-rc.2) (2021-09-20)

[Full Changelog](https://github.com/kubeflow/tf-operator/compare/v1.3.0-rc.1...v1.3.0-rc.2)

**Closed issues:**

- \[bug\] Unable to specify pod template metadata for TFJob [\1403](https://github.com/kubeflow/tf-operator/issues/1403)

**Merged pull requests:**

- Bump controller-tools to 0.6.0 and enable GenerateEmbeddedObjectMeta [\1410](https://github.com/kubeflow/tf-operator/pull/1410) ([Jeffwan](https://github.com/jeffwan))
- chore: Update image tags in manifests [\1412](https://github.com/kubeflow/tf-operator/pull/1412) ([Jeffwan](https://github.com/jeffwan))

[v1.3.0-rc.1](https://github.com/kubeflow/tf-operator/tree/v1.3.0-rc.1) (2021-09-15)

[Full Changelog](https://github.com/kubeflow/tf-operator/compare/v1.3.0-rc.0...v1.3.0-rc.1)

**Closed issues:**

- \[bug\] Reconcilation fails when upgrading common to 0.3.6 [\1394](https://github.com/kubeflow/tf-operator/issues/1394)

**Merged pull requests:**

- Update manifests with latest image tag [\1406](https://github.com/kubeflow/tf-operator/pull/1406) ([johnugeorge](https://github.com/johnugeorge))
- 2010: fix to expose correct monitoring port [\1405](https://github.com/kubeflow/tf-operator/pull/1405) ([deepak-muley](https://github.com/deepak-muley))
- Fix 1399: added pod matching label in service selector [\1404](https://github.com/kubeflow/tf-operator/pull/1404) ([deepak-muley](https://github.com/deepak-muley))
- fix: runPolicy validation error in the examples [\1401](https://github.com/kubeflow/tf-operator/pull/1401) ([Jeffwan](https://github.com/Jeffwan))

[v1.3.0-rc.0](https://github.com/kubeflow/tf-operator/tree/v1.3.0-rc.0) (2021-08-31)

[Full Changelog](https://github.com/kubeflow/tf-operator/compare/v1.2.0...v1.3.0-rc.0)

Features

* add api-doc for all frameworks ([1370](https://github.com/kubeflow/tf-operator/pull/1370), DeliangFan)
* Update training operator release process ([1347](https://github.com/kubeflow/tf-operator/pull/1347), Jeffwan)
* add option to enable schemes ([1342](https://github.com/kubeflow/tf-operator/pull/1342), zw0610)
* Don't use inline for runPolicy and register defaulter function ([1330](https://github.com/kubeflow/tf-operator/pull/1330), Jeffwan)
* Use low-level controller and handlers in SetupWithManager ([1315](https://github.com/kubeflow/tf-operator/pull/1315), Jeffwan)
* Reorganize controllers code structure ([1302](https://github.com/kubeflow/tf-operator/pull/1302), Jeffwan)
* Unify code structure of training job api ([1300](https://github.com/kubeflow/tf-operator/pull/1300), Jeffwan)
* Extract reusable codes to common utility ([1297](https://github.com/kubeflow/tf-operator/pull/1297), Jeffwan)
* Add MXJob v1 api and controller ([1296](https://github.com/kubeflow/tf-operator/pull/1296), Jeffwan)
* add tf reconciler to the unified operator ([1295](https://github.com/kubeflow/tf-operator/pull/1295), zw0610)
* Add XGBoost controller ([1293](https://github.com/kubeflow/tf-operator/pull/1293), Jeffwan)
* add pytorch API and controller ([1294](https://github.com/kubeflow/tf-operator/pull/1294), zw0610)
* Generate tfjob 1.19.x compatible clientset ([1290](https://github.com/kubeflow/tf-operator/pull/1290), Jeffwan)
* add XGBoostJob api ([1286](https://github.com/kubeflow/tf-operator/pull/1286), zw0610)

Bug fixes

* fix: volcano pod group creation issue ([1390](https://github.com/kubeflow/tf-operator/pull/1390), Jeffwan)
* Fix 1340 prometheus counters ([1375](https://github.com/kubeflow/tf-operator/pull/1375), deepak-muley)
* fix makefile to store crds in a separate folder ([1368](https://github.com/kubeflow/tf-operator/pull/1368), deepak-muley)
* Fix copyright header for some files ([1371](https://github.com/kubeflow/tf-operator/pull/1371), DeliangFan)
* fix incorrect torch env population ([1361](https://github.com/kubeflow/tf-operator/pull/1361), Jeffwan)
* fix: Resolve scheme registration issue for defaulters ([1360](https://github.com/kubeflow/tf-operator/pull/1360), Jeffwan)
* Fix XGBoost container name in log message ([1362](https://github.com/kubeflow/tf-operator/pull/1362), andreyvelich)
* Fix postsubmit job using PULL_BASE_SHA ([1344](https://github.com/kubeflow/tf-operator/pull/1344), Jeffwan)
* Fix all client request that needs contexts ([1292](https://github.com/kubeflow/tf-operator/pull/1292), Jeffwan)

Misc

* Add simple verification jobs ([1391](https://github.com/kubeflow/tf-operator/pull/1391), Jeffwan)
* chore: Bump kubeflow/common version to 0.3.7 ([1388](https://github.com/kubeflow/tf-operator/pull/1388), Jeffwan)
* chore(doc): Update README.md ([1387](https://github.com/kubeflow/tf-operator/pull/1387), Jeffwan)
* Remove lagacy tf-operator from the codebase ([1378](https://github.com/kubeflow/tf-operator/pull/1378), thunderboltsid)
* chore: Update manifest image tag ([1364](https://github.com/kubeflow/tf-operator/pull/1364), Jeffwan)
* add example for mxnet and pytorch ([1373](https://github.com/kubeflow/tf-operator/pull/1373), DeliangFan)
* docs: Update to use kubectl kustomize in instructions ([1356](https://github.com/kubeflow/tf-operator/pull/1356), terrytangyuan)
* Clean up manifests and remove unused files ([1349](https://github.com/kubeflow/tf-operator/pull/1349), Jeffwan)
* 1322: Modified manifests to use all-in-one training-operator ([1346](https://github.com/kubeflow/tf-operator/pull/1346), deepak-muley)
* chore: Update changelog.md ([1339](https://github.com/kubeflow/tf-operator/pull/1339), Jeffwan)
* Move root level docs to docs folder ([1338](https://github.com/kubeflow/tf-operator/pull/1338), Jeffwan)
* Temporarily add Jeffwan to OWNERS ([1287](https://github.com/kubeflow/tf-operator/pull/1287), Jeffwan)

Testing

* Update include dirs in prow config ([1374](https://github.com/kubeflow/tf-operator/pull/1374), andreyvelich)
* Enable e2e test against universal operator ([1336](https://github.com/kubeflow/tf-operator/pull/1336), Jeffwan)
* Use PULL_PULL_SHA for image ([1334](https://github.com/kubeflow/tf-operator/pull/1334), PatrickXYS)