harshraj11584 / Human-Activity-Recognition-0.97-Kaggle

Human Activity Recognition UCI Dataset, final score 0.97196

Human-Activity-Recognition-Kaggle-0.97

Human Activity Recognition UCI Dataset [link], final score 0.97196

[Competition Link]

d_nn.ipynb first finds features with high correlation, drops them, then uses Sequential DNN for prediction.
Accuracy = 0.95443

Predictor.ipynb uses xgboost for classification, for different values of number of estimators and max depth.
Accuracy = 0.97196