Bot releases are hidden (Show)
kfp
a required dependency of AutoMLOpskfp<2.0.0
was required. kfp v2 is not always backwards compatibile with v1AutoMLOps.component
and AutoMLOps.pipeline
syntax rather than dsl.component
AutoMLOps.component
be python primitives only.AutoMLOps-cache
; files are no longer written to an itermediary location and stored in memory insteadPublished by srastatter 8 months ago
setup_model_monitoring
parameter to AutoMLOps.generate
and orchestration/configs.pycreate_model_monitoring_job.sh
script to generateAutoMLOps.monitor
function, along with relevant templates and testsconfig/defaults.yaml
filelogging.googleapis.com
to the list of potentially required apis.AutoMLOps.monitor
: get_model_monitoring_min_permissions
and get_model_monitoring_recommended_roles
in utils.pyservices/submission_service/main.py.j2
to include elements for automatic retraining based on monitoring anomaly logs, and adding in labels to the submit.services/submission_service/requirements.txt
to include google-cloud-storage.README.md.j2
to reflect the optional creation of the new model_monitoring/ directory.config/defaults.yaml
file gets written; this file is now written using the write_yaml_file
function (yaml.safe_dump) in utils.py.account_permissions_warning
function in utils.py to include a new operation: operation='model_monitoring'
validate_schedule
in utils.py to validate_use_ci
to reflect new requirements for model monitoring.Published by srastatter 9 months ago
Published by srastatter 12 months ago
Published by srastatter 12 months ago
automlops
branchPublished by srastatter about 1 year ago
Published by srastatter about 1 year ago
Published by srastatter about 1 year ago
Published by srastatter about 1 year ago
Published by srastatter about 1 year ago
use_ci=False
.use_ci=False
provision directory not being created properlyuse_ci=False
terraform configPublished by srastatter about 1 year ago
Major version updates:
Published by srastatter about 1 year ago
Published by srastatter over 1 year ago
kfp<2.0.0.
to address the recent migration to kfp2+.Published by srastatter over 1 year ago
clear_cache
function which deletes all files within the tmpfiles directory.use_kfp_spec
from parameter lists and switched to determining this at run-time.Published by srastatter over 1 year ago
Published by srastatter over 1 year ago
Published by srastatter over 1 year ago
Official release on PyPI.
Published by srastatter over 1 year ago
Added feature to allow for accelerated/distributed model training.
Published by srastatter over 1 year ago
Reworked process to submit jobs to cloud runner service.
Published by srastatter over 1 year ago
Official release of AutoMLOps!
AutoMLOps is a service that generates a production ready MLOps pipeline from Jupyter Notebooks, bridging the gap between Data Science and DevOps and accelerating the adoption and use of Vertex AI. The service generates an MLOps codebase for users to customize, and provides a way to build and manage a CI/CD integrated MLOps pipeline from the notebook. AutoMLOps automatically builds a source repo for versioning, cloudbuild configs and triggers, an artifact registry for storing custom components, gs buckets, service accounts and updated IAM privs for running pipelines, enables APIs (cloud Run, Cloud Build, Artifact Registry, etc.), creates a runner service API in Cloud Run for submitting PipelineJobs to Vertex AI, and a Cloud Scheduler job for submitting PipelineJobs on a recurring basis. These automatic integrations empower data scientists to take their experiments to production more quickly, allowing them to focus on what they do best: providing actionable insights through data.