augerai/a2ml

Issues

• Actuals should be sent to endpoint only if multipredict_enabled is TRUE

  #602 opened 3 years ago by skatedplrn
  1

• Support experiments metadata for MLRAM

  #597 opened 3 years ago by holyketzer
  0

• Method to build/get metrics for the trial

  #586 opened 3 years ago by skatedplrn
  0

• Support predict with downloaded model

  #559 opened 3 years ago by skatedplrn
  0

• Support API for R language

  #579 opened 3 years ago by skatedplrn
  0

• Support multi-predict

  #571 opened 3 years ago by holyketzer
  0

• feature_divergence models should be recreated with new sklearn

  #562 opened 3 years ago by skatedplrn
  0

• Build docker with azure

  #561 opened 3 years ago by skatedplrn
  0

• Add baseline_scores to actual endpoint

  #551 opened 3 years ago by skatedplrn
  0

• Check how and/or works in top expression

  #546 opened 4 years ago by holyketzer
  0

• A2ML is not able to report about SIGKILL during task handling

  #530 opened 4 years ago by holyketzer
  0

• Ability to use A2ML from Jupyter Notebook

  #529 opened 4 years ago by holyketzer
  0

• ROI: support predefined formulas with parameters

  #503 opened 4 years ago by skatedplrn
  0

• ExternalModel: make sure review command return enough information to run retrain

  #497 opened 4 years ago by skatedplrn
  0

• MLRAM: review command should return as much information as possible

  #511 opened 4 years ago by skatedplrn
  0

• FP and FN counts for multi class problems

  #504 opened 4 years ago by holyketzer
  0

• BusinessValue: new metrics for regression

  #496 opened 4 years ago by skatedplrn
  0

• Tasks which handle datasets should be run in separate process

  #495 opened 4 years ago by skatedplrn
  0

• send providers to create project file

  #494 opened 4 years ago by skatedplrn
  0

• Iill-defined empirical covariance in build diverge model task

  #484 opened 4 years ago by holyketzer
  1

• Support actuals for project

  #468 opened 4 years ago by skatedplrn
  1

• Support external models in A2ML CLI

  #480 opened 4 years ago by holyketzer
  0

• Incorporate the review-demo tool into a2ml as an e2e testing system

  #422 opened 4 years ago by knobotz
  1

• SHAP feature importance can give a nice charts

  #479 opened 4 years ago by holyketzer
  0

• remove datetime column from distribution chart

  #457 opened 4 years ago by skatedplrn
  0

• support postgres url as data source

  #459 opened 4 years ago by skatedplrn
  2

• Feature importance doesn't return anything for ensembles

  #456 opened 4 years ago by holyketzer
  3

• Add date_field_name to actuals

  #458 opened 4 years ago by skatedplrn
  0

• Auger credentials for A2ML in staging and production

  #462 opened 4 years ago by holyketzer
  1

• Azure: support algorithms search space

  #411 opened 4 years ago by skatedplrn
  0

• Support review with actuals only

  #435 opened 4 years ago by skatedplrn
  0

• Add getting feature importance for models

  #432 opened 4 years ago by holyketzer
  1

• Enable tag-based deploys to a2ml production

  #384 opened 4 years ago by knobotz
  0

• Support Hub endpoints and review API

  #354 opened 4 years ago by holyketzer
  3

• Auger cloud user cluster parameter

  #396 opened 4 years ago by skatedplrn
  1

• Use source_format for auger cloud

  #381 opened 4 years ago by skatedplrn
  0

• Document experiment/metric for providers

  #412 opened 4 years ago by skatedplrn
  0

• Azure: support metrics

  #410 opened 4 years ago by skatedplrn
  0

• update_cluster_config_task doesn't appear to interpret params correctly

  #375 opened 4 years ago by knobotz
  0

• ModelReview.add_actuals() performs dataframe merge that consumes too much memory on large datasets

  #395 opened 4 years ago by knobotz
  1

• build_review_data_task should save output file in parquet format

  #370 opened 4 years ago by holyketzer
  0

• Issue with update_cluster_config_task on prod

  #391 opened 4 years ago by holyketzer
  0

• DatasetValidationError on import_data_task in Azure

  #373 opened 4 years ago by holyketzer
  2

• High CPU usage with distribution_chart_stats_task

  #360 opened 4 years ago by skatedplrn
  0

• feature importance for azure in distribution_chart_stats task

  #361 opened 4 years ago by skatedplrn
  0

• New s3fs release (0.5.0) breaks our usage of boto

  #362 opened 4 years ago by knobotz
  0

• distribution_chart_stats task should return "actual_<target>" and "predicted_<target>"

  #358 opened 4 years ago by skatedplrn
  0

• A2ML.predict call should allow a prediction ID to be passed to it

  #350 opened 4 years ago by adamblum
  1

• make delete_actuals synchronous

  #352 opened 4 years ago by skatedplrn
  0

• r2 metric for actual return > 1

  #347 opened 4 years ago by skatedplrn
  3