BTB
BTB copied to clipboard
A simple, extensible library for developing AutoML systems
Sometimes there are some hyperparameter combinations that provoke errors, so we know that those are bad values, but we don't have any score to pass to the tuner. Should we...
Some classifiers accept multiple types for the same HyperParameter. An example of this is the `max_features` hyperparameter of the [RandomForestRegressor](http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestRegressor.html#sklearn.ensemble.RandomForestRegressor): > **max_features** : int, float, string or None, optional (default="auto")...
There are some cases where a hyperparameter is a range of values. An example of this is the `quantile_range` in [scikit-learn Robust Scaler](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.RobustScaler.html) Currently, a user could separate this range...
The tuner provides a high-level propose/add API, in which it maintains an internal state of parameters that have been tried and the associated score, and updates its state in response...
Integrate with [Codacy](https://app.codacy.com/projects?/login) for automatic analysis/commenting on PR's for code quality and style. Would reduce a lot of the work in analyzing PR's.
* baytune 0.5.0 * macOS 14.1 * python 3.8 The following line produces an error when `scikit-learn New in version 1.2: `sparse` was renamed to `sparse_output` https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html
Resolve broken tests `test___init___default_values` and `test___init___users_values` from `TestBaseMetaModelTuner`. At this moment, they are commented out.