ML Report

1 Data Pre-Processing and Feature Engineering -

1.1 Regression

## Recipe
## 
## Inputs:
## 
##       role #variables
##    outcome          1
##  predictor          6
## 
## Training data contained 29701 data points and no missing data.
## 
## Operations:
## 
## Zero variance filter removed <none> [trained]
## Yeo-Johnson transformation on temperature [trained]
## Centering for traffic [trained]
## Scaling for traffic [trained]
## Dummy variables from Time, weather_general, weather_detailed [trained]

1.1.1 Splitting main data into train and test

1.1.1.1 Train Dataset

1.1.1.2 Test Dataset

1.2 Classification

## Recipe
## 
## Inputs:
## 
##       role #variables
##    outcome          1
##  predictor         19
## 
## Training data contained 1500 data points and no missing data.
## 
## Operations:
## 
## Dummy variables from age, gender, education, country, ethnicity, consumption_a... [trained]

1.2.1 Splitting main data into train and test

1.2.1.1 Train Dataset

1.2.1.2 Test Dataset

2 Modelling

2.1 Regression

2.1.1 Model_01_linear_glmnet_ridge

2.1.2 Model_02_linear_glmnet_lasso

2.1.3 Model_03_linear_glmnet_elastic

2.1.4 Model_04_boost_tree_xgboost

2.1.5 Model_05_svm_rbf

2.2 Classification

2.2.1 Random Forest (Ranger)

2.2.2 Random Forest (Rand)

2.2.3 XGBoost

3 Applying future Dataset

3.1 Regression

3.1.1 Model_01_linear_glmnet_ridge

3.1.2 Model_02_linear_glmnet_lasso

3.1.3 Model_03_linear_glmnet_elastic

3.1.4 Model_04_boost_tree_xgboost

3.1.5 Model_05_svm_rbf

3.2 Classification

3.2.1 Random Forest (Ranger)

3.2.2 Random Forest (Rand)

3.2.3 XGBoost