search

Colin-Codes / IntentClassifier-ML-Project

Pyhton, Keras, SciKit-Learn, Matplotlib: Machine learning research project around classification of intent behind tech support emails in order to enable automatic follow up.
0 stars 0 forks source link

How to improve cross validation by selecting folds #28

Open Colin-Codes opened 4 years ago

Colin-Codes commented 4 years ago

https://stats.stackexchange.com/questions/27730/choice-of-k-in-k-fold-cross-validation

Colin-Codes commented 4 years ago
  1. Optimisation of cross validation: I'm confident I can optimise cross validation folds by looking for the smoothest curve on graphs of the k parameter of the K-Nearest Neighbours classifier versus accuracy, when varying the number of folds between graphs. Although a more mathematical approach would be preferable. I have struggled to find instances of cross validation for neural networks - do you know of any good examples?