def run_pipeline(request, pipeline_metadata, pipeline_package_path=None,
pipeline_id=None):
"""Run a pipeline."""
client = _get_client(pipeline_metadata.get("kfp_host", None))
experiment = client.create_experiment(pipeline_metadata["experiment_name"])
run_name = kfp_utils.generate_run_name(pipeline_metadata["pipeline_name"])
run = client.run_pipeline(experiment.id, run_name,
pipeline_package_path=pipeline_package_path,
pipeline_id=pipeline_id)
return {"id": run.id, "name": run.name, "status": run.status}
results_task = results_op()\
.add_pvolumes(pvolumes_dict)\
.after(randomforest_task, logisticregression_task, naivebayes_task, svm_task, decisiontree_task)
results_task.container.working_dir = "/kale"
results_task.container.set_security_context(
k8s_client.V1SecurityContext(run_as_user=0))
output_artifacts = {}
output_artifacts.update(
{'mlpipeline-ui-metadata': '/mlpipeline-ui-metadata.json'})
output_artifacts.update({'results': '/results.html'})
results_task.output_artifact_paths.update(output_artifacts)
if __name__ == "__main__":
pipeline_func = auto_generated_pipeline
pipeline_filename = pipeline_func.__name__ + '.pipeline.tar.gz'
import kfp.compiler as compiler
compiler.Compiler().compile(pipeline_func, pipeline_filename)
# Get or create an experiment and submit a pipeline run
import kfp
client = kfp.Client()
experiment = client.create_experiment('titanic')
# Submit a pipeline run
from kale.utils.kfp_utils import generate_run_name
run_name = generate_run_name('titanic-ml-rnd')
run_result = client.run_pipeline(experiment.id, run_name,
pipeline_filename, {})
{'mlpipeline-ui-metadata': '/mlpipeline-ui-metadata.json'})
output_artifacts.update({'explain': '/explain.html'})
explain_task.output_artifact_paths.update(output_artifacts)
explain_task.add_pod_label("pipelines.kubeflow.org/metadata_written",
"true")
dep_names = explain_task.dependent_names + volume_step_names
explain_task.add_pod_annotation("kubeflow-kale.org/dependent-templates",
json.dumps(dep_names))
if volume_name_parameters:
explain_task.add_pod_annotation(
"kubeflow-kale.org/volume-name-parameters",
json.dumps(volume_name_parameters))
if __name__ == "__main__":
pipeline_func = auto_generated_pipeline
pipeline_filename = pipeline_func.__name__ + '.pipeline.tar.gz'
import kfp.compiler as compiler
compiler.Compiler().compile(pipeline_func, pipeline_filename)
# Get or create an experiment and submit a pipeline run
import kfp
client = kfp.Client()
experiment = client.create_experiment('seldon-e2e-adult')
# Submit a pipeline run
from kale.utils.kfp_utils import generate_run_name
run_name = generate_run_name('seldon-e2e-adult-ttonn')
run_result = client.run_pipeline(experiment.id, run_name,
pipeline_filename, {})
{'mlpipeline-ui-metadata': '/mlpipeline-ui-metadata.json'})
output_artifacts.update({'results': '/results.html'})
results_task.output_artifact_paths.update(output_artifacts)
results_task.add_pod_label(
"pipelines.kubeflow.org/metadata_written", "true")
dep_names = results_task.dependent_names + volume_step_names
results_task.add_pod_annotation(
"kubeflow-kale.org/dependent-templates", json.dumps(dep_names))
if volume_name_parameters:
results_task.add_pod_annotation(
"kubeflow-kale.org/volume-name-parameters",
json.dumps(volume_name_parameters))
if __name__ == "__main__":
pipeline_func = auto_generated_pipeline
pipeline_filename = pipeline_func.__name__ + '.pipeline.tar.gz'
import kfp.compiler as compiler
compiler.Compiler().compile(pipeline_func, pipeline_filename)
# Get or create an experiment and submit a pipeline run
import kfp
client = kfp.Client()
experiment = client.create_experiment('titanic')
# Submit a pipeline run
from kale.utils.kfp_utils import generate_run_name
run_name = generate_run_name('titanic-ml-gxj28')
run_result = client.run_pipeline(
experiment.id, run_name, pipeline_filename, {})
{'mlpipeline-ui-metadata': '/mlpipeline-ui-metadata.json'})
output_artifacts.update({'sum_matrix': '/sum_matrix.html'})
sum_matrix_task.output_artifact_paths.update(output_artifacts)
sum_matrix_task.add_pod_label("pipelines.kubeflow.org/metadata_written",
"true")
dep_names = sum_matrix_task.dependent_names + volume_step_names
sum_matrix_task.add_pod_annotation("kubeflow-kale.org/dependent-templates",
json.dumps(dep_names))
if volume_name_parameters:
sum_matrix_task.add_pod_annotation(
"kubeflow-kale.org/volume-name-parameters",
json.dumps(volume_name_parameters))
if __name__ == "__main__":
pipeline_func = auto_generated_pipeline
pipeline_filename = pipeline_func.__name__ + '.pipeline.tar.gz'
import kfp.compiler as compiler
compiler.Compiler().compile(pipeline_func, pipeline_filename)
# Get or create an experiment and submit a pipeline run
import kfp
client = kfp.Client()
experiment = client.create_experiment('hp-tuning')
# Submit a pipeline run
from kale.utils.kfp_utils import generate_run_name
run_name = generate_run_name('hp-test-rnd')
run_result = client.run_pipeline(experiment.id, run_name,
pipeline_filename, {})
