Mar-27-2019, 04:22 PM
Hey all,
Six months ago a friend and I got started on this interesting challenge: Could one accurately predict the training time of common data science algorithms such as Random Forest, Svm or Kmeans?
Our python package called “Scitime” (https://github.com/nathan-toubiana/scitime) (which you can pip or conda install) is the result of our effort to build a scalable solution that can be applied to any Scikit learn algorithms in the future.
We detailed our methodology and findings in this article: https://medium.freecodecamp.org/two-hour...cf755df938
Try it out and let us know what you think!
Six months ago a friend and I got started on this interesting challenge: Could one accurately predict the training time of common data science algorithms such as Random Forest, Svm or Kmeans?
Our python package called “Scitime” (https://github.com/nathan-toubiana/scitime) (which you can pip or conda install) is the result of our effort to build a scalable solution that can be applied to any Scikit learn algorithms in the future.
We detailed our methodology and findings in this article: https://medium.freecodecamp.org/two-hour...cf755df938
Try it out and let us know what you think!