Skip to content

Version 2.0.0b1 #892

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
renesass merged 473 commits into from
Jan 7, 2023
Merged

Version 2.0.0b1 #892

renesass merged 473 commits into from
Jan 7, 2023

Conversation

@renesass
Copy link
Collaborator

@renesass renesass commented Dec 7, 2022 •
edited
Loading

This version contains primarily the new intensifier changes.

Documentation

https://automl.github.io/SMAC3/development/

Required for release candidate

  • Intensifier rework:
    • Support for multi-threating
    • Consider trials in runhistory
  • SH/HB
    • Easier implementation
    • Consider trials in runhistory
  • Fix multi-tasking on macs
  • Update changelog
  • Update docs
  • Specify some docstrings (inconsistent API in runhistory)
  • Reach 87% test coverage
  • Benchmarking

Intensifier

  • Decide which trial to suggest next if we are waiting for all busy workers (try new config instead of new instance, seed pair of already evaluated config)
  • Should directly operate on the RunHistory (no process_results method anymore)
  • When should the model be trained? static approach: after n evaluated configs
  • If user provided instances, seeds in the RunHistory: Those are preferred to be evaluated first
  • get_next_config: Generator, yields a config. Count how many configs yielded inside the generator.
  • Intensifier: Generator, yield trial.
  • Get rid off t_intensify (does not work in multi-processing) -- what was the idea behind it? Make sure that the computation time of the surrogate model and the intensifier are balanced?

renesass and others added 30 commits May 5, 2022 09:45
@renesass
References to BOHB
58176fc
@benjamc
Make format (isort)
76abc8c
@renesass
Fix tests?
3099206
@renesass
Updated MO docs
d0f9fe2
@dengdifan
test and examples for EIPS acquisition function
8ca230e
@renesass
Merge branch 'main' into development
581a0b6
@dengdifan
fix mypy
beda941
@dengdifan
pre-comit
16aab39
@dengdifan
add docstrings
09e79d6
@BastianZim
Delete .DS_Store
d8bc619
@BastianZim
Ignore .DS_Store
6d90d39
@renesass
Hotfix: Update cached costs for MO
9d238c3
@renesass
Merge branch 'development' into BastianZim/main
873db36
@renesass
Merge pull request #838 from BastianZim/main
b92e7f0 
Ignore and delete .DS_Store
@renesass
Updated changelog
985c07f
@renesass
Updated version
a6f81dd
@renesass
Added entry point
98f063e
@renesass
isort+black
18ee262
@DE0CH
change README
32e8d54
@dengdifan
update example
c562497
@dengdifan
pre-commit
d0219b5
@dengdifan
Revert "pre-commit"
caeee1b 
This reverts commit d0219b5.
@dengdifan
Revert "avoid incrementing twice when computing EIPS"
d6a1f54 
This reverts commit 31d2381.
@dengdifan
fix rh2epm eips computation
025c7b2
@dengdifan
rf with multi-objective
5957363
@dengdifan
Revert "rf with multi-objective"
14904e0 
This reverts commit ba85d1f.
@dengdifan
avoid incrementing twice when computing EIPS
573f7a6
@dengdifan
multi-objective surrogate model
cecd4af
@dengdifan
test and examples for EIPS acquisition function
7702f91
@dengdifan
fix mypy
7aa9977
@renesass renesass added this to the v2.0.0 milestone Dec 22, 2022
@renesass renesass added enhancement documentation Documentation is needed/added. feature test Tests are needed/added. example labels Dec 22, 2022
@renesass
Copy link
Collaborator Author

renesass commented Dec 30, 2022

Okay guys, I just made my last pass and fixed some last minute issues. Please have a look and confirm ASAP.

@renesass renesass self-assigned this Dec 30, 2022
benjamc
benjamc approved these changes Jan 5, 2023
View reviewed changes
tests/test_model/test_rf.py
# targets = np.log(y)
# model = RandomForest(
# configspace=_get_cs(X.shape[1]),
# ratio_features=1.0,
Copy link
Collaborator

@benjamc benjamc Jan 5, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Commented out bc should be tested in the RF repo? (Same for GP?)

Copy link
Collaborator Author

@renesass renesass Jan 7, 2023

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

it's because Boston is not supported anymore... changing the dataset means changing all of the values

@renesass renesass merged commit 0a1dff4 into main Jan 7, 2023
github-actions bot pushed a commit that referenced this pull request Jan 7, 2023
René Sass: Version 2.0.0b1 (#892)
c263312
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@benjamc benjamc benjamc approved these changes

@dengdifan dengdifan Awaiting requested review from dengdifan

@helegraf helegraf Awaiting requested review from helegraf

@timruhkopf timruhkopf Awaiting requested review from timruhkopf

+1 more reviewer

@sarah-segel sarah-segel sarah-segel approved these changes

Reviewers whose approvals may not affect merge requirements

Assignees

@renesass renesass

Labels

documentation Documentation is needed/added. enhancement example feature test Tests are needed/added.

Projects

SMAC board
Status: Done

Milestone

v2.0.0

Development

Successfully merging this pull request may close these issues.