Experiment Set 1 - Dummy Classifier

Description: In this set, a dummy classifier is utilized, where a decision tree is constructed based on level 1 preprocessed features. Level 1 preprocessing refers to minimal preprocessing required for the algorithm to function.

Experiment Set 2 - XGBoost Without FE

Description: This set employs the XGBoost algorithm without any feature engineering (FE) techniques.

Experiment Set 3 - XGBoost With FE

Description: XGBoost is utilized in this set with feature engineering techniques applied to enhance model performance.

Experiment Set 4 - XGBoost With FE & Optuna

Description: XGBoost is employed with feature engineering techniques, and Optuna is used for hyperparameter optimization.

Experiment Set 5 - LightGBM Without FE

Description: LightGBM is utilized in this set without any feature engineering.

Experiment Set 6 - LightGBM With FE

Description: LightGBM is employed with feature engineering techniques to improve model performance.

Experiment Set 7 - LightGBM With FE & Optuna

Description: LightGBM is used with feature engineering techniques, and Optuna is employed for hyperparameter tuning.

Experiment Set 8 - LightGBM With custom loss function (focal loss)

Description: This set utilizes LightGBM with a custom loss function, specifically the focal loss function, to address class imbalance or prioritize hard-to-classify samples.

Experiment Set 9 - Catboost With FE

Description: CatBoost algorithm is employed with feature engineering techniques.

Experiment Set 10 - Catboost With FE & Optuna

Description: CatBoost algorithm is utilized with feature engineering techniques, and Optuna is employed for hyperparameter optimization.

Experiment Set 11 - Pycaret

Description: This set utilizes the Pycaret library, which automates the machine learning workflow, including data preprocessing, model selection, and evaluation.

Experiment Set 12 - Autogluon

Description: Autogluon is utilized in this set, which is an automated machine learning library designed for easy-to-use and efficient model selection and hyperparameter tuning.

Experiment Set 13 - AutoXGB

Description: AutoXGB employs automated methods, likely through techniques such as AutoML, for training and optimizing an XGBoost model.

Experiment Set 14 - AutoLGBM

Description: AutoLGBM utilizes automated techniques, possibly through AutoML, for training and optimizing a LightGBM model.

Experiment Set 15 - Advanced LightGBM

Description: This set employs advanced techniques or configurations with LightGBM to enhance model performance further.

Experiment Set 16 - XGB Stacked

Description: In this set, an ensemble approach is used where XGBoost models are stacked to improve overall predictive performance.

Experiment Set 17 - Neural Network

Description: This set utilizes a neural network-based approach for modeling, which can handle complex relationships in the data.

Experiment Set 18

Description: (Fill in the description of Experiment Set 18)

Experiment Set 19

Description: (Fill in the description of Experiment Set 19)

Experiment Set 20 - XGB with NN probabilities

Description: This set combines the predictions of an XGBoost model with probabilities obtained from a neural network model, likely for ensemble or boosting purposes.