auto-sklearn
auto-sklearn copied to clipboard
automl
Automated Machine Learning with scikit-learn
Note to anyone who reads this issue, please feel free to leave comments on what you would like to see included. # Adding detailed documentation Currently the documentation lacks an...
# Short Question Description Why does the implementation of Ensemble Selection not return the best ensemble found during the greedy search? Following the algorithm presented in Caruana et al. (2004),...
Initialization of hyperparameter search space fails when the first preprocessor method can not be used with none of the estimators. Example: Using in the fit method: ``` include_preprocessors = ['kernel_pca','select_rates']...
The problem I want to use auto-sklearn on is a time-series. Can we modify sklearn to include cv with time series?
This is mostly a discussion item. PR #1402 unveiled a bug in the handling of pandas that could lead to drastically worse performance (see [here](https://github.com/automl/auto-sklearn/runs/5200936217?check_suite_focus=true)). It was uncovered because the...
FastICA currently looks for `n_components` between 10 and 2000, with a default of 100. However, assuming there are more rows N than columns M, a data set can only support...
As I have learned the metalearning feature has about 30 fixed configurations to try out. How many configuration should a autosklearn fit task try out as a rule of thumb?...
Because `fit` optimizes the machine learning models and builds an ensemble out of them, it cannot be used for basic sklearn functions such as `cross_val_predict`. https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.cross_val_predict.html `cross_val_predict` would call `fit`...
In the implementation of balancing, whether to use class weight in init param or sample weight in fit param depends on hard coded lists of classifiers. Currently only decision tree...
I'm working on a project that needs me to pass a mask of label into the scorer. For example, if y is a vector (100,), then I will have a...
