Open utterances-bot opened 3 years ago
I think your tutorials are absolutely fantastic! I have just one question: How would you perform feature selection using tidymodels? For instance: In the field of proteomics and genetics expression data of genes or proteins are used for the prediction of certain diseases. These data sets often consists of the expression data of thousands of genes or proteins. What would the tidymodels approach be to select the best ones to build your model?
There is already support for some approaches to feature selection in tidymodels (you can check out, for example, how to use lasso regression or something like this approach for evaluating different predictor sets) but if there is a specific approach that is interesting to your use case, let us know and we can look into that!
+1 for a tutorial/demo of tidymodels elasticnet with 'omic data, though it may just be lasso with grid tuning of the mixture parameter (I think..)
Learn tidymodels with my supervised machine learning course | Julia Silge
I am happy to announce that a new version of my free, online, interactive course has been published!
https://juliasilge.com/blog/tidymodels-ml-course/