hyper parameters collection

Support Vector Machines (SVM)

• Kernel: Type of kernel function to use (linear, polynomial, radial basis function, etc.)

• C: Penalty term for misclassification

• Gamma: Kernel coefficient

Random Forest

• N_estimators: Number of trees in the forest

• Max_depth: Maximum depth of each tree

• Min_samples_split: Minimum number of samples required to split a node

• Min_samples_leaf: Minimum number of samples required at a leaf node

Logistic Regression

• C: Inverse of regularization strength

• Penalty: Type of regularization (l1, l2, etc.)

• Solver: Algorithm to use for optimization

K-Nearest Neighbors (KNN)

• N_neighbors: Number of nearest neighbors to consider

• Weights: Weight function to use for prediction (uniform, distance-based, etc.)

• Algorithm: Algorithm to use for searching for neighbors (brute force, kd-tree, etc.)

• 
  

Naive Bayes

• Prior: Prior probabilities for each class

Decision Trees

• Criterion: Measure of quality of a split (gini, entropy)

• Max_depth: Maximum depth of the tree

• Min_samples_split: Minimum number of samples required to split a node

• Min_samples_leaf: Minimum number of samples required at a leaf node

Neural Networks

• Activation function: Function to use for activation in hidden layers (sigmoid, tanh, ReLU, etc.)

• Learning rate: Rate at which the model updates weights during training

• Hidden layers: Number and size of hidden layers

• Batch size: Number of samples per gradient update

• Epochs: Number of iterations over the entire dataset during training

AdaBoost

• Base estimator: Base model to use for boosting

• N_estimators: Number of boosting iterations

• Learning rate: Rate at which the model updates weights during training

Gradient Boosting

• Loss function: Loss function to use for optimization

• N_estimators: Number of boosting iterations

• Learning rate: Rate at which the model updates weights during training

• Max_depth: Maximum depth of each tree

XGBoost

• Learning rate: Rate at which the model updates weights during training

• N_estimators: Number of boosting iterations

• Max_depth: Maximum depth of each tree

• Colsample_bytree: Subsample ratio of columns when constructing each tree

• Subsample: Subsample ratio of rows when constructing each tree

short compilation

1. Logistic Regression - penalty (L1 or L2), C (regularization strength)

2. Decision Tree - max_depth, min_samples_split, min_samples_leaf

3. Random Forest - n_estimators, max_depth, min_samples_split, min_samples_leaf

4. Support Vector Machine - kernel (linear, rbf, poly), C (regularization strength)

5. Naive Bayes - alpha (smoothing parameter)

6. K-Nearest Neighbors - n_neighbors, metric (distance measure)

7. Gradient Boosting - n_estimators, learning_rate, max_depth

8. XGBoost - n_estimators, learning_rate, max_depth, gamma (minimum loss reduction)

9. Neural Network - hidden_layer_sizes, activation (function), solver (optimization algorithm)

10. AdaBoost - n_estimators, learning_rate

11. Bagging - n_estimators, max_samples, max_features

12. LightGBM - learning_rate, num_leaves, min_child_samples

13. CatBoost - depth, learning_rate, l2_leaf_reg

14. Extreme Gradient Boosting (XGBoost) - n_estimators, learning_rate, max_depth, min_child_weight

15. Ridge Regression - alpha (regularization strength), solver (optimization algorithm)