Skip to content

Feature engineering+feature interaction+ANN to top the chart of ML kaggle competition

Notifications You must be signed in to change notification settings

raizen74/Kaggle-NN

Repository files navigation

Kaggle-NN

Given 900000 instances and 31 features representing manufacturing control data, build a neural network to predict whether the machine is in state 0 or state 1.

  • 51% of the machines are in state 0 and 49% in state 1.

My model achieved 97,4% accuracy prediction on test data, topping the chart of the ML kaggle competition.

About

Feature engineering+feature interaction+ANN to top the chart of ML kaggle competition

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published