skll
skll copied to clipboard
SciKit-Learn Laboratory (SKLL) makes it easy to run machine learning experiments.
This PR adds a simple utility method to the Learner class, allowing it to return the feature names used at the estimator step as a 1-D array. Learners with a...
Learners with a selector potentially have a smaller feature name size than is stored in the feature_vectorizer. A simple method accessing the actual used feature names for a learner with...
When you specify a new `predict` method for a custom learner, the interaction with the skll `Learner` method of the same name is unclear. For example, the `Learner predict` method...
Looks like scikit-learn is planning to remove the boston housing dataset in v1.2 because it has ethical issues. They recommend using the calfornia housing dataset instead. We should do this...
We should try and match the way that scikit-learn writes their docstrings. [Here](https://github.com/scikit-learn/scikit-learn/blob/master/sklearn/linear_model/_huber.py#L125) is an example. For example, there is a blank line between the parameters and the default values...
I know this will be disagreeable to some (including me actually!), but it would be a good idea to at least think about this. `black` is nice, even if aggressive.
This PR adds two new learners `OrdinalRidge` and `LAD` from [mord library ](https://pythonhosted.org/mord/index.html). There are two things I would like to mention here: - These learners do not have `rescaled`...
The mord package provides some options: https://pythonhosted.org/mord/