python_backend/src/CodeRunAndModelTrain/config/model_train.yaml

classification_models:
  LogisticRegression:
    class: LogisticRegression
    default_params:
      penalty: l2
      C: 1.0
      solver: lbfgs
      max_iter: 100
      multi_class: auto
      random_state: None

  KNeighborsClassifier:
    class: KNeighborsClassifier
    default_params:
      n_neighbors: 5
      weights: uniform
      algorithm: auto
      p: 2
      n_jobs: None

  SVC:
    class: SVC
    default_params:
      C: 1.0
      kernel: rbf
      gamma: scale
      degree: 3
      probability: False
      random_state: None

  LinearSVC:
    class: LinearSVC
    default_params:
      C: 1.0
      loss: squared_hinge
      penalty: l2
      dual: True
      max_iter: 1000
      random_state: None

  DecisionTreeClassifier:
    class: DecisionTreeClassifier
    default_params:
      criterion: gini
      max_depth: None
      min_samples_split: 2
      min_samples_leaf: 1
      random_state: None

  RandomForestClassifier:
    class: RandomForestClassifier
    default_params:
      n_estimators: 100
      criterion: gini
      max_depth: None
      min_samples_split: 2
      min_samples_leaf: 1
      bootstrap: True
      random_state: None
      n_jobs: None

  GradientBoostingClassifier:
    class: GradientBoostingClassifier
    default_params:
      loss: log_loss # or 'deviance' in older versions
      learning_rate: 0.1
      n_estimators: 100
      subsample: 1.0
      max_depth: 3
      random_state: None

  AdaBoostClassifier:
    class: AdaBoostClassifier
    default_params:
      n_estimators: 50
      learning_rate: 1.0
      algorithm: SAMME.R
      random_state: None

  ExtraTreesClassifier:
    class: ExtraTreesClassifier
    default_params:
      n_estimators: 100
      criterion: gini
      max_depth: None
      min_samples_split: 2
      min_samples_leaf: 1
      bootstrap: False
      random_state: None
      n_jobs: None

  GaussianNB:
    class: GaussianNB
    default_params:

  MultinomialNB:
    class: MultinomialNB
    default_params:
      alpha: 1.0
      fit_prior: True

  BernoulliNB:
    class: BernoulliNB
    default_params:
      alpha: 1.0
      fit_prior: True

  MLPClassifier:
    class: MLPClassifier
    default_params:
      hidden_layer_sizes: (100)
      activation: relu
      solver: adam
      alpha: 0.0001
      learning_rate: constant
      max_iter: 200
      random_state: None

  SGDClassifier:
    class: SGDClassifier
    default_params:
      loss: hinge # linear SVM by default
      penalty: l2
      alpha: 0.0001
      learning_rate: optimal
      max_iter: 1000
      random_state: None

  Perceptron:
    class: Perceptron
    default_params:
      penalty: None
      alpha: 0.0001
      max_iter: 1000
      random_state: None

  PassiveAggressiveClassifier:
    class: PassiveAggressiveClassifier
    default_params:
      C: 1.0
      max_iter: 1000
      random_state: None

  RidgeClassifier:
    class: RidgeClassifier
    default_params:
      alpha: 1.0
      fit_intercept: True
      solver: auto
      random_state: None

# ================ Regression ===============

regression_models:
  LinearRegression:
    class: LinearRegression
    default_params:
      fit_intercept: True
      copy_X: True
      n_jobs: None

  Ridge:
    class: Ridge
    default_params:
      alpha: 1.0
      fit_intercept: True
      solver: auto
      random_state: None

  Lasso:
    class: Lasso
    default_params:
      alpha: 1.0
      fit_intercept: True
      max_iter: 1000
      random_state: None

  ElasticNet:
    class: ElasticNet
    default_params:
      alpha: 1.0
      l1_ratio: 0.5
      fit_intercept: True
      max_iter: 1000
      random_state: None

  KNeighborsRegressor:
    class: KNeighborsRegressor
    default_params:
      n_neighbors: 5
      weights: uniform
      algorithm: auto
      p: 2
      n_jobs: None

  SVR:
    class: SVR
    default_params:
      kernel: rbf
      C: 1.0
      epsilon: 0.1
      gamma: scale
      degree: 3

  LinearSVR:
    class: LinearSVR
    default_params:
      epsilon: 0.0
      C: 1.0
      loss: epsilon_insensitive
      max_iter: 1000
      random_state: None

  DecisionTreeRegressor:
    class: DecisionTreeRegressor
    default_params:
      criterion: squared_error
      max_depth: None
      min_samples_split: 2
      min_samples_leaf: 1
      random_state: None

  RandomForestRegressor:
    class: RandomForestRegressor
    default_params:
      n_estimators: 100
      criterion: squared_error
      max_depth: None
      min_samples_split: 2
      min_samples_leaf: 1
      bootstrap: True
      random_state: None
      n_jobs: None

  ExtraTreesRegressor:
    class: ExtraTreesRegressor
    default_params:
      n_estimators: 100
      criterion: squared_error
      max_depth: None
      min_samples_split: 2
      min_samples_leaf: 1
      bootstrap: False
      random_state: None
      n_jobs: None

  GradientBoostingRegressor:
    class: GradientBoostingRegressor
    default_params:
      loss: squared_error
      learning_rate: 0.1
      n_estimators: 100
      subsample: 1.0
      max_depth: 3
      random_state: None

  AdaBoostRegressor:
    class: AdaBoostRegressor
    default_params:
      n_estimators: 50
      learning_rate: 1.0
      loss: linear
      random_state: None

  MLPRegressor:
    class: MLPRegressor
    default_params:
      hidden_layer_sizes: (100)
      activation: relu
      solver: adam
      alpha: 0.0001
      learning_rate: constant
      max_iter: 200
      random_state: None

  SGDRegressor:
    class: SGDRegressor
    default_params:
      loss: squared_error
      penalty: l2
      alpha: 0.0001
      learning_rate: invscaling
      max_iter: 1000
      random_state: None