Skip to content

Releases: MLBazaar/BTB

v0.5.0

28 Jul 16:53
9b4500e
Compare
Choose a tag to compare

What's Changed

New Contributors

Full Changelog: v0.4.0...v0.5.0

0.4.0 - 2020-12-30

30 Dec 19:20
Compare
Choose a tag to compare

This release increases the supported version of python to 3.8 and also includes changes in the
installation requirements, where pandas and scikit-optimize packages have been updated
to support higher versions.

Internal improvements

  • Added github actions.

Resolved Issues

  • Issue #210: Integrate Scikit-Optimize for benchmarking.

Benchmarking

Screenshot from 2020-12-30 19-42-23

0.3.12 - 2020-09-08

08 Sep 10:25
Compare
Choose a tag to compare

In this release BTB includes two new tuners, based on Copulas : GCP and GCPEi. They use a GaussianCopulaProcessRegressor meta-model which uses sklearn.gaussian_process after applying copulas.univariate.Univariate transformations to the input data. The data is afterwards reverted when making a proposal for the hyperparameters .

Resolved Issues

  • Issue #15: Implement a GaussianCopulaProcessRegressor.
  • Issue #205: Separate datasets from MLChallenge.
  • Issue #208: Implement GaussianCopulaProcessMetaModel.

Benchmarking

Benchmarking Results

Detailed results from which this summary emerged are available here.

0.3.11 - 2020-06-12

12 Jun 14:41
Compare
Choose a tag to compare

With this release we fix the AX.optimize tuning function by casting the values of the hyperparameters to the type of value that they represent.

Resolved Issues

  • Issue #201: Fix AX.optimize malfunction.

0.3.10 - 2020-05-29

29 May 21:51
Compare
Choose a tag to compare

With this release we integrate a new tuning library, SMAC, with our benchmarking process. A new
leaderboard including this library has been generated. The following two tuners from this library
have been added:

  • SMAC4HPO: Bayesian optimization using a Random Forest model of pyrfr.
  • HB4AC: Uses Successive Halving for proposals.

Internal improvements

  • Renamed btb_benchmark/tuners to btb_benchmark/tuning_functions.
  • Ready to use tuning functions from btb_benchmark/tuning_functions.

Resolved Issues

  • Issue #195: Integrate SMAC for benchmarking.

0.3.9 - 2020-05-18

18 May 18:38
Compare
Choose a tag to compare

With this release we integrate a new tuning function for the benchmarking process: Ax.optimize, that implements Bayesian optimization and bandit optimization, powered by BoTorch. A new leaderboard including this tuning function has been generated.

0.3.8 - 2020-05-08

08 May 19:08
Compare
Choose a tag to compare

This version adds a new functionality which allows running the benchmarking framework on a Kubernetes cluster. By doing this, the benchmarking process can be executed distributedly, which reduces the time necessary to generate a new leaderboard.

Internal improvements

  • btb_benchmark.kubernetes.run_dask_function: Run dask function inside a pod using the given config.
  • btb_benchmark.kubernetes.run_on_kubernetes: Start a Dask Cluster using dask-kubernetes and run a function.
  • Documentation updated.
  • Jupyter notebooks with examples on how to run the benchmarking process and how to run it on kubernetes.

0.3.7 - 2020-04-15

15 Apr 11:41
Compare
Choose a tag to compare

This release brings a new benchmark framework with public leaderboard.
As part of our benchmarking efforts we will run the framework at every release and make the results
public. In each run we compare it to other tuners and optimizer libraries. We are constantly adding
new libraries for comparison. If you have suggestions for a tuner library we should include in our
compraison, please contact us via email at [email protected].

Resolved Issues

  • Issue #159: Implement more MLChallenges and generate a public leaderboard.
  • Issue #180: Update BTB Benchmarking module.
  • Issue #182: Integrate HyperOPT with benchmarking.
  • Issue #184: Integrate dask to benchmarking.

0.3.6 - 2020-03-04

04 Mar 09:09
Compare
Choose a tag to compare

This release improves BTBSession error handling and allows Tunables with cardinality
equal to 1 to be scored with BTBSession. Also, we provide a new documentation for
this version of BTB.

Internal Improvements

Improved documentation, unittests and integration tests.

Resolved Issues

  • Issue #164: Improve documentation for v0.3.5+.
  • Issue #166: Wrong erro raised by BTBSession on too many errors.
  • Issue #170: Tuner has no scores attribute until record is run once.
  • Issue #175: BTBSession crashes when record is not performed.
  • Issue #176: BTBSession fails to select a proper Tunable when normalized_scores becomse None.

0.3.5 - 2020-01-21

21 Jan 17:51
Compare
Choose a tag to compare

With this release we are improving BTBSession by adding private attributes, or not intended to
be public / modified by the user and also improving the documentation of it.

Internal Improvements

Improved docstrings, unittests and public interface of BTBSession.

Resolved Issues

  • Issue #162: Fix session with the given comments on PR 156.