scikit-learn-mooc
scikit-learn-mooc copied to clipboard
INRIA
Machine learning in Python with scikit-learn MOOC
Extracting information from a given step inside a pipeline can be tricky and is not explained -yet used- along the MOOC (See [this post in the forum](https://mooc-forums.inria.fr/moocsl/t/coefficients-in-pipeline/4793)). One possible solution...
- [x] 1. add a new notebook between the notebook 1 and 2 to introduce the model evaluation. It should discuss `train_test_split` and introduce `cross_validate`. This would be a notebook...
Split from #398: The proposal of Arturo is good: - [x] move score distribution and variations at the beginning of the module. covered by adding the video in M1: https://inria.github.io/scikit-learn-mooc/predictive_modeling_pipeline/02_numerical_pipeline_video_cross_validation.html...
A bit related to #439, [this forum comment](https://mooc-forums.inria.fr/moocsl/t/thank-you-for-the-great-course/4467/3) suggest adding videos for cross-validation methods. This could be done with a single video either as an introduction to M7 or as...
The first time we use a train-validation-test split for parameter tuning is at the beginning of M3 in [this exercise](https://inria.github.io/scikit-learn-mooc/python_scripts/parameter_tuning_sol_02.html), as we (silently) use CV within the training set and...
Adding an overview of model families was proposed [here in the forum](https://mooc-forums.inria.fr/moocsl/t/end-of-course-comment/4748). > I understand it is difficult to have focus on all type of modelling methods. Nevertheless I feel...
Box-plots are frequently used during the MOOC but a person without a minimal formation in statistics might not understand how to read them. I think the soon-to-be-added notebook on score...
Most of the notebooks use `ColumnTransformer` glossary.md predictive_modeling_module_take_away.md: wrap_up_quiz 1 03_categorical_pipeline_column_transformer.py 03_categorical_pipeline_ex_02.py 03_categorical_pipeline_sol_02.py parameter_tuning_ex_02.py and parameter_tuning_sol_02.py parameter_tuning_ex_03.py and parameter_tuning_sol_03.py parameter_tuning_grid_search.py parameter_tuning_nested.py parameter_tuning_randomized_search.py On the other hand, `make_column_transformer` is only used...
I wrote an example in scikit-learn which I think would be useful in the inspection part of the course: https://github.com/scikit-learn/scikit-learn/pull/18821
Now that the hyper-parameter tuning has been moved out of the modelling predictive pipeline part, we do: - adult_census for Modeling Predictive Pipeline - boston_housing for Selecting the Best Model...