gravatar

FrankyCardinale

Francesco Cardinale

Recently Published

Stacking Mean LR & XGB - Pred & Perc Error
Stacking Mean Linear Regressione & XGBoost - Pred & Perc Error
Stacking Mean LR & XGB - Perc Error
Stacking with mean of Linear Regressione and XGBOOST
Comparison of test data sets and subset (Normalized with std dev) giving the worst cases for all model
Comparison of test data sets and subset giving the worst cases for all model: - datasets (SNS with 0) normalized by dividing variables by the std dev of m1, .., m11 - Worst Cases: Error > 50% for all models
Model Comparison - Normalized dataset (v2)
Normalization of datasets by dividing variables by the standard deviation of m1, … , m11
HTML
Comparison of test data sets and subset (Normalized) giving the worst cases for all model
Comparison of test data sets and subset giving the worst cases for all model - datasets (SNS with 0) normalized by dividing variables by the max of m1, .., m11 - Worst Cases: Error > 50% for all models
Model Comparison - Normalized dataset
Normalization of datasets by dividing variables by the max of m1, .., m11
Comparison of data sets giving the worst cases for each of four model (SNS with 0)
Comparison of data sets giving the worst cases (Perc Error > 50) for each of four model (SNS with 0)
Model Stacking - Mean of the best 3 Models (SNS with 0)
Mean of the best 3 models on dataset with 0 - H2O XGBOOST - Linear Regression - Random Forest
Model Stacking - Mean of the best 3 Models (SNS without 0)
Mean of the best 3 models on dataset without 0 - Linear Regression - Random Forest Regression - Gradient Boosted Trees Regression
Naïve Models - SNS Without 0
Model 1: Pred = mean(m1 ... m11) Model 2: Pred = m11
Naïve Models - SNS With 0
Model 1: Pred = mean(m1 ... m11) Model 2: Pred = m11