v0.3.0

• Search space improvement
• Separation of initial points and low cost partial config in HPO input

v0.2.9

A number of new features and improvements:

• pickle the AutoML object
• get best model per estimator
• bug fix
• add verbose argument in AutoML.fit()
• nni support
• stateless HPO API

v0.2.8

perf improvement for parallel tuning and bug fix

v0.2.6

added support for nested search space
added experimental support for nni
minor bug fix

v0.2.4

make sample_weight right for ensemble; reduced dependency

v0.2.3

minor bug fix for rf

v0.2.2

Enhanced algorithms and implementations of hyperparameter tuning with support for parallel tuning and ray tune;
Support keyword arguments like sample_weight;
Enable mlflow logging.

v0.1.3

Set default logging level to INFO

v0.1.2

Hyperparameter + sample size: Cost Frugal Optimization (CFO) + Estimated Cost of Improvement (ECI)
Learner selection: ECI
Resampling strategy: threshold rule

v0.1.0

First release of FLAML.

Hyperparameter + sample size: Cost Frugal Optimization (CFO) + Estimated Cost of Improvement (ECI)
Learner selection: ECI
Resampling strategy: threshold rule