TFjob v1 launcher #2677
There are no files selected for viewing
File renamed without changes.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,4 @@ | ||
| approvers: | ||
| - hougangliu | ||
| reviewers: | ||
| - hougangliu |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,35 @@ | ||
| name: Kubeflow - Launch TFJob | ||
| description: Kubeflow TFJob launcher | ||
| inputs: | ||
| - {name: Name, type: String, description: 'TFJob name.'} | ||
| - {name: Namespace, type: String, default: kubeflow, description: 'TFJob namespace.'} | ||
| - {name: Version, type: String, default: v1, description: 'TFJob version.'} | ||
| - {name: ActiveDeadlineSeconds, type: Integer, default: -1, description: 'Specifies the duration (in seconds) since startTime during which the job can remain active before it is terminated. Must be a positive integer. This setting applies only to pods where restartPolicy is OnFailure or Always.'} | ||
| - {name: BackoffLimit, type: Integer, default: -1, description: 'Number of retries before marking this job as failed.'} | ||
| - {name: ttl Seconds After Finished, type: Integer, default: -1, description: 'Defines the TTL for cleaning up finished TFJobs.'} | ||
| - {name: CleanPodPolicy, type: String, default: Running, description: 'Defines the policy for cleaning up pods after the TFJob completes.'} | ||
| - {name: PS Spec, type: JSON, default: '{}', description: 'TFJob ps replicaSpecs.'} | ||
| - {name: Worker Spec, type: JSON, default: '{}', description: 'TFJob worker replicaSpecs.'} | ||
| - {name: Chief Spec, type: JSON, default: '{}', description: 'TFJob chief replicaSpecs.'} | ||
| - {name: Evaluator Spec, type: JSON, default: '{}', description: 'TFJob evaluator replicaSpecs.'} | ||
| - {name: Tfjob Timeout Minutes, type: Integer, default: 1440, description: 'Time in minutes to wait for the TFJob to complete.'} | ||
| - {name: Delete Finished Tfjob, type: Bool, default: 'True' , description: 'Whether to delete the tfjob after it is finished.'} | ||
| implementation: | ||
| container: | ||
| image: liuhougangxa/kubeflow-tfjob-launcher:latest | ||
| command: [python, /ml/launch_tfjob.py] | ||
| args: [ | ||
| --name, {inputValue: Name}, | ||
| --namespace, {inputValue: Namespace}, | ||
| --version, {inputValue: Version}, | ||
| --activeDeadlineSeconds, {inputValue: ActiveDeadlineSeconds}, | ||
| --backoffLimit, {inputValue: BackoffLimit}, | ||
| --cleanPodPolicy, {inputValue: CleanPodPolicy}, | ||
| --ttlSecondsAfterFinished, {inputValue: ttl Seconds After Finished}, | ||
| --psSpec, {inputValue: PS Spec}, | ||
| --workerSpec, {inputValue: Worker Spec}, | ||
| --chiefSpec, {inputValue: Chief Spec}, | ||
| --evaluatorSpec, {inputValue: Evaluator Spec}, | ||
| --tfjobTimeoutMinutes, {inputValue: Tfjob Timeout Minutes}, | ||
| --deleteAfterDone, {inputValue: Delete Finished Tfjob}, | ||
| ] |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,75 @@ | ||
| import json | ||
| from kfp import components | ||
| import kfp.dsl as dsl | ||
|
|
||
| @dsl.pipeline( | ||
| name="Launch kubeflow tfjob", | ||
| description="An example to launch tfjob." | ||
| ) | ||
| def mnist_train( | ||
| name="mnist", | ||
| namespace="kubeflow", | ||
| workerNum=3, | ||
| ttlSecondsAfterFinished=-1, | ||
| tfjobTimeoutMinutes=60, | ||
| deleteAfterDone=False): | ||
| tfjob_launcher_op = components.load_component_from_file("./component.yaml") | ||
| # tfjob_launcher_op = components.load_component_from_url('https://raw.githubusercontent.com/kubeflow/pipelines/master/components/kubeflow/launcher/component.yaml') | ||
|
There was a problem hiding this comment. JFYI, the link already exists (use your commit hash instead of master). |
||
|
|
||
| chief = { | ||
| "replicas": 1, | ||
| "restartPolicy": "OnFailure", | ||
| "template": { | ||
| "spec": { | ||
| "containers": [ | ||
| { | ||
| "command": [ | ||
| "python", | ||
| "/opt/model.py" | ||
| ], | ||
| "args": [ | ||
| "--tf-train-steps=6000" | ||
| ], | ||
| "image": "liuhougangxa/tf-estimator-mnist", | ||
| "name": "tensorflow", | ||
| } | ||
| ] | ||
| } | ||
| } | ||
| } | ||
| worker = {} | ||
| if workerNum > 0: | ||
| worker = { | ||
| "replicas": workerNum, | ||
| "restartPolicy": "OnFailure", | ||
| "template": { | ||
| "spec": { | ||
| "containers": [ | ||
| { | ||
| "command": [ | ||
| "python", | ||
| "/opt/model.py" | ||
| ], | ||
| "args": [ | ||
| "--tf-train-steps=6000" | ||
| ], | ||
| "image": "liuhougangxa/tf-estimator-mnist", | ||
| "name": "tensorflow", | ||
| } | ||
| ] | ||
| } | ||
| } | ||
| } | ||
| tfjob_launcher_op( | ||
| name=name, | ||
| namespace=namespace, | ||
| ttl_seconds_after_finished=ttlSecondsAfterFinished, | ||
| worker_spec=worker, | ||
| chief_spec=chief, | ||
| tfjob_timeout_minutes=tfjobTimeoutMinutes, | ||
| delete_finished_tfjob=deleteAfterDone | ||
| ) | ||
|
|
||
| if __name__ == "__main__": | ||
| import kfp.compiler as compiler | ||
| compiler.Compiler().compile(mnist_train, __file__ + ".tar.gz") | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should probably switch to python3 at some point.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think so