Release v0.1.9rc3

[bouthilx]

Important changes

New plotting API

Plotting capability is being added to experiment clients. You can now plot the regret (curve of best objective found during optimization) with simply experiment.plot.regret(). You can find an example here.

Web API (REST server)

A web API was added for v0.1.9 in order to support the visualization dashboard that is currently under work. See full documentation here.

Database commands rm & set

Command line helpers have been added to simplify the process of deleting experiments and trials as well as modifying trials in the database. See full documentation for both commands here.

Detailed list of changes

New features

• New plotting API (Plotting API Starter #400, Analysis module #426)
• REST server (Adding trials/ REST resource #437, REST plotting endpoint with regret plots #438, Documenting the full specification of the web api #441, Refactoring REST responses #446)
• New command orion db rm (Add command orion db rm #425)
• New command orion db set (Add command orion db set #428)

Breaking changes

• Drop support for orion < 0.1.6 (Drop support for orion < 0.1.6 #420)
• Deprecate init_only and add hunt --init-only (Deprecate init_only and add hunt --init-only #436)

Bug Fixes

• Fix backward compatibility tests in master - Drop support for orion < 0.1.6 (Drop support for orion < 0.1.6 #420)
• Fix test using 'in' to assert equality (Fix test using 'in' to assert equality #427)
• Avoid side effects in setup_storage (Avoid side effects in setup_storage #444)
• Return onehot interval instead of categorical (Return onehot interval instead of categorical #447)
• Adjust OneHotEncoding interval shape (Adjust OneHotEncoding interval shape #449)
• Support non existing user script conf files (Support non existing user script conf files  #452)
• Support load yaml config for dict items (Support load yaml config for dict items #453)
• Fix fetch_lost_trials (Causing random FailedUpdate when using many workers) (Print user friendly err msg if script inexecutable #445)

Other improvements

• Improve CI - Convert functional algo tests to python API (Convert functional algo tests to python API #418)
• Monitor code change between trial execution (Quit if code changed between two trials execution #429)
• Add info about working_dir when using orion info CLI (added info about working_dir when using orion info CLI #430)
• Enable is_broken for ExperimentView (Enable is_broken for ExperimentView #432)
• Specialized properties for Trials (Specialized properties for Trials #433)
• Add completed field to info cmd (Add completed field to info cmd #435)
• Add orion.client.get_experiment() (Add orion.client.get_experiment() #439)
• Sort modules and extensions alphabetically (Sort modules and extensions alphabetically #443)
• Print user friendly err msg if script inexecutable (Print user friendly err msg if script inexecutable #445)
• Migrate from Travis to Github-actions (Migrate from Travis to github actions #458, Update test badge #459, Rework Githuh Actions into a single workflow #461)

Documentation improvements

• Add a code of conduct (Code of conduct #353)
• Refresh database documentation (Refresh database documentation #413)
• Merging Sequential and Service API (Merging Sequential and Service API #416)
• CLI documentation improvements (CLI documentation improvements #440)
• Remove outdated branch_from parameter documentation in experiment_builder.build() (Remove outdated branch_from parameter documentation in experiment_builder.build() #442)
• Changed variable name in docs to reflect the correct name (changed variable name in docs to reflect the correct name #448)
• Fix config of EvolutionaryES (Fix config of EvolutionaryES #451)

[codecov]

Codecov Report

Merging #462 (1b3921c) into master (d192c02) will increase coverage by 1.81%.
The diff coverage is 73.34%.

@@            Coverage Diff             @@
##           master     #462      +/-   ##
==========================================
+ Coverage   47.52%   49.33%   +1.81%     
==========================================
  Files          70       81      +11     
  Lines       13009    13922     +913     
  Branches      322      345      +23     
==========================================
+ Hits         6183     6869     +686     
- Misses       6800     7036     +236     
+ Partials       26       17       -9     

Flag	Coverage Δ
backward	97.81% <100.00%> (?)
unittests	48.82% <72.97%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/orion/core/cli/__init__.py	38.88% <ø> (-50.00%)	⬇️
tests/unittests/core/cli/test_info.py	19.92% <0.00%> (-0.30%)	⬇️
tests/unittests/core/worker/test_trial.py	17.64% <7.01%> (ø)
tests/unittests/core/test_transformer.py	21.38% <8.33%> (-0.29%)	⬇️
tests/unittests/storage/test_storage.py	16.58% <12.65%> (-0.95%)	⬇️
tests/unittests/core/worker/test_experiment.py	17.11% <13.33%> (-0.12%)	⬇️
tests/unittests/core/io/test_config.py	16.04% <18.75%> (+0.29%)	⬆️
tests/unittests/core/io/test_experiment_builder.py	16.92% <19.04%> (+0.12%)	⬆️
...sts/unittests/core/io/test_orion_cmdline_parser.py	20.21% <21.21%> (+0.86%)	⬆️
tests/unittests/plotting/test_plotly_backend.py	31.37% <31.37%> (ø)
... and 44 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d192c02...1b3921c. Read the comment docs.