GoogleCloudPlatform/ml-testing-accelerators issues and pull requests

#100 - Move GCS bucket names to cluster-level ConfigMap.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#100 - Move GCS bucket names to cluster-level ConfigMap.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#99 - Update PyTorch models to use PersistentVolumeClaim

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#98 - Dashboard: use download command from BQ if it exists

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#98 - Dashboard: use download command from BQ if it exists

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#97 - Alert email tz

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#97 - Alert email tz

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#96 - Feature request: more configurable email alerts

Issue - State: open - Opened by zcain117 over 4 years ago

#96 - Feature request: more configurable email alerts

Issue - State: open - Opened by zcain117 over 4 years ago

#95 - Overwrite cache when running HF tests

Pull Request - State: closed - Opened by jysohn23 over 4 years ago - 4 comments

#95 - Overwrite cache when running HF tests

Pull Request - State: closed - Opened by jysohn23 over 4 years ago - 4 comments

#94 - Add in pod convergence tests for classifier_trainner models.

Pull Request - State: closed - Opened by allenwang28 over 4 years ago - 5 comments

#94 - Add in pod convergence tests for classifier_trainner models.

Pull Request - State: closed - Opened by allenwang28 over 4 years ago - 5 comments

#93 - Update classifier tests to only override epochs.

Pull Request - State: closed - Opened by allenwang28 over 4 years ago

#92 - Fix import error in dashboard.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#92 - Fix import error in dashboard.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#91 - Convert timestamp to US/Pacific before using the date in dashboard.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#91 - Convert timestamp to US/Pacific before using the date in dashboard.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#90 - Remove pytorch-1.5 automated tests.

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 2 comments

#90 - Remove pytorch-1.5 automated tests.

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 2 comments

#89 - Dashboard unit testing

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#89 - Dashboard unit testing

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#88 - Add region option to accelerator specs.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#88 - Add region option to accelerator specs.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#87 - Correct the final convergence accuracy values for classifier models

Pull Request - State: closed - Opened by allenwang28 over 4 years ago

#87 - Correct the final convergence accuracy values for classifier models

Pull Request - State: closed - Opened by allenwang28 over 4 years ago

#86 - Add build args to TensorFlow Dockerfile.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#86 - Add build args to TensorFlow Dockerfile.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#85 - Add gcloud back to TF image.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#85 - Add gcloud back to TF image.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#84 - Remove tests for old ResNet compile/fit model.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#84 - Remove tests for old ResNet compile/fit model.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#83 - Store workload link and logs download command in bigquery.

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 1 comment

#83 - Store workload link and logs download command in bigquery.

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 1 comment

#82 - Add support for tagging failures in dashboard

Issue - State: open - Opened by will-cromar over 4 years ago - 2 comments
Labels: enhancement

#82 - Add support for tagging failures in dashboard

Issue - State: open - Opened by will-cromar over 4 years ago - 2 comments
Labels: enhancement

#81 - Update RetineNet model parameters

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#81 - Update RetineNet model parameters

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#80 - Stop trying to delete old rows from Bigquery.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#80 - Stop trying to delete old rows from Bigquery.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#79 - Dashboard initial

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 1 comment

#79 - Dashboard initial

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 1 comment

#78 - Dashboard showing one line error msgs for failed tests could be useful.

Issue - State: open - Opened by taylanbil over 4 years ago
Labels: enhancement

#78 - Dashboard showing one line error msgs for failed tests could be useful.

Issue - State: open - Opened by taylanbil over 4 years ago
Labels: enhancement

#77 - Dashboard colors get stuck when a cell is clicked.

Issue - State: closed - Opened by taylanbil over 4 years ago - 2 comments

#77 - Dashboard colors get stuck when a cell is clicked.

Issue - State: closed - Opened by taylanbil over 4 years ago - 2 comments

#76 - Dashboard test names are not Ctrl+f searchable

Issue - State: open - Opened by jysohn23 over 4 years ago - 1 comment

#76 - Dashboard test names are not Ctrl+f searchable

Issue - State: open - Opened by jysohn23 over 4 years ago - 1 comment

#75 - Rename 'functional' and 'convergence' in metrics handler.

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#75 - Rename 'functional' and 'convergence' in metrics handler.

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#74 - Skip PubSub publish step for one-shot jobs.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#74 - Skip PubSub publish step for one-shot jobs.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#73 - Fix regression_test_config for HF tests.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#73 - Fix regression_test_config for HF tests.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#72 - Add tests for Huggingface GLUE tasks

Pull Request - State: closed - Opened by jysohn23 over 4 years ago

#72 - Add tests for Huggingface GLUE tasks

Pull Request - State: closed - Opened by jysohn23 over 4 years ago

#71 - Set maximum character limit on test names

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#71 - Set maximum character limit on test names

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#70 - Publish PubSub message from an initContainer

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#70 - Publish PubSub message from an initContainer

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#69 - Set 500 steps_per_loop for TF BERT SQuAD

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#69 - Set 500 steps_per_loop for TF BERT SQuAD

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#68 - Delete the redundant last bucket check in the GCS-checkpointing test.

Pull Request - State: closed - Opened by taylanbil over 4 years ago

#68 - Delete the redundant last bucket check in the GCS-checkpointing test.

Pull Request - State: closed - Opened by taylanbil over 4 years ago

#67 - Change deployment manager config to template with optional pools.

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#67 - Change deployment manager config to template with optional pools.

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 1 comment

#66 - Added FairSeq checkpointing tests, to local disk and gcs.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 4 comments

#66 - Added FairSeq checkpointing tests, to local disk and gcs.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 4 comments

#65 - Oneshot retries may be unnecessary.

Issue - State: closed - Opened by taylanbil over 4 years ago - 1 comment
Labels: enhancement

#65 - Oneshot retries may be unnecessary.

Issue - State: closed - Opened by taylanbil over 4 years ago - 1 comment
Labels: enhancement

#64 - Lower retry limit and job history limit

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#64 - Lower retry limit and job history limit

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#63 - Fix classifier tests

Pull Request - State: closed - Opened by allenwang28 over 4 years ago

#63 - Fix classifier tests

Pull Request - State: closed - Opened by allenwang28 over 4 years ago

#62 - Add mixin for experimental tests.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#62 - Add mixin for experimental tests.

Pull Request - State: closed - Opened by will-cromar over 4 years ago

#61 - Handle duplicate rows in Bigquery.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#61 - Handle duplicate rows in Bigquery.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#60 - Set msg_publish_time column. Use None temporarily.

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 1 comment

#59 - Add scripts to re-gen yamls.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 2 comments

#59 - Add scripts to re-gen yamls.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 2 comments

#58 - Increase efficientnet epochs to 500 to fix convergence issues.

Pull Request - State: closed - Opened by allenwang28 over 4 years ago - 1 comment

#58 - Increase efficientnet epochs to 500 to fix convergence issues.

Pull Request - State: closed - Opened by allenwang28 over 4 years ago - 1 comment

#57 - Updating Keras API repository

Pull Request - State: closed - Opened by gagika over 4 years ago - 2 comments

#57 - Updating Keras API repository

Pull Request - State: closed - Opened by gagika over 4 years ago - 2 comments

#56 - Add experimental PyTorch MNIST TPU pod test

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 5 comments

#56 - Add experimental PyTorch MNIST TPU pod test

Pull Request - State: closed - Opened by will-cromar over 4 years ago - 5 comments

#55 - Make resnet50-mp target lower case.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 3 comments

#54 - Handle case where k8s conditions are None for non-success jobs.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#54 - Handle case where k8s conditions are None for non-success jobs.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#53 - Fix pytorch test targets to use lowercase filename.

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 1 comment

#53 - Fix pytorch test targets to use lowercase filename.

Pull Request - State: closed - Opened by zcain117 over 4 years ago - 1 comment

#52 - Update templates README to point to metrics regression config documen…

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#52 - Update templates README to point to metrics regression config documen…

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#51 - Change pytorch transformer convergence test to using BLEU score, and fast config.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 9 comments

#51 - Change pytorch transformer convergence test to using BLEU score, and fast config.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 9 comments

#50 - Lowercase the testnames for Resnet-MP.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 1 comment

#50 - Lowercase the testnames for Resnet-MP.

Pull Request - State: closed - Opened by taylanbil over 4 years ago - 1 comment

#49 - Update pytorch-1.5 roberta test to use newer parameterization.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

#49 - Update pytorch-1.5 roberta test to use newer parameterization.

Pull Request - State: closed - Opened by zcain117 over 4 years ago

GitHub / GoogleCloudPlatform/ml-testing-accelerators issues and pull requests