Metrics
Metrics copied to clipboard
Machine learning evaluation metrics, implemented in Python, R, Haskell, and MATLAB / Octave
1. when actual list zero lists, the apk should alwasy return 1.0; The current versio of code will always return 0.0 which is different from the code of function using...
Small update in style of enumerate
I came across this problem when I tried to use ap@k for a recommendation system that returns no items if the confidence is low.
Hello, is there some preferred way, such as readily built command, to plot the AUC curve with Metrics package such as with RORC pkg [1]? [1] https://stats.stackexchange.com/questions/132777/what-does-auc-stand-for-and-what-is-it
not actual may have problem with numpy
I want to test the ap@K in recommender system. Can I use it to achieve it with the matrix U ,V and test data.
What should the input format to the function mapk be? (ie, in what format should actual and predicted be?)