Content:

  1. Installing nnetsauce for Python
  2. Classification
  3. Regression

Disclaimer: I have no affiliation with the lazypredict project.

A few days ago, I stumbled accross a cool Python package called lazypredict. Pretty well-designed, working, and relying on scikit-learn’s design.

With lazypredict, you can rapidly have an idea of which scikit-learn model (can also work with xgboost’s and lightgbm’s scikit-learn-like interfaces) performs best on a given data set, with a little preprocessing, and without hyperparameters’ tuning (this is important to note).

I thought something similar could be beneficial to nnetsauce’s classes CustomClassifier, CustomRegressor (see detailed examples below, and interact with the graphs) and MTS. For now.

So far, in nnetsauce (Python version), I adapted the lazy prediction feature to regression (CustomRegressor) and classification (CustomClassifier). Not for univariate and multivariate time series forecasting (MTS) yet. You can try it from a GitHub branch.

2 - Installation

!pip install git+https://github.com/Techtonique/nnetsauce.git@lazy-predict

2 - Classification

2 - 1 Loading the Dataset

import nnetsauce as ns
from sklearn.datasets import load_breast_cancer

data = load_breast_cancer()
X = data.data
y= data.target

2 - 2 Building the classification model using LazyPredict

from sklearn.model_selection import train_test_split

# split the data
X_train, X_test, y_train, y_test = train_test_split(X, y,
                                                    test_size=0.2,
                                                    random_state=123)

# build the lazyclassifier
clf = ns.LazyClassifier(verbose=0, ignore_warnings=True,
                        custom_metric=None,
                        n_hidden_features=10,
                        col_sample=0.9)

# fit it
models, predictions = clf.fit(X_train, X_test, y_train, y_test)

100%|██████████| 27/27 [00:09<00:00,  2.71it/s]

# print the best models
display(models)
Accuracy Balanced Accuracy ROC AUC F1 Score Time Taken
Model
LogisticRegression 0.99 0.99 0.99 0.99 0.69
LinearSVC 0.98 0.98 0.98 0.98 0.33
SGDClassifier 0.98 0.98 0.98 0.98 0.19
Perceptron 0.98 0.98 0.98 0.98 0.15
LabelPropagation 0.98 0.98 0.98 0.98 0.33
LabelSpreading 0.98 0.98 0.98 0.98 0.43
SVC 0.98 0.98 0.98 0.98 0.16
RandomForestClassifier 0.98 0.98 0.98 0.98 0.66
ExtraTreesClassifier 0.98 0.98 0.98 0.98 0.40
KNeighborsClassifier 0.98 0.98 0.98 0.98 0.34
DecisionTreeClassifier 0.97 0.97 0.97 0.97 0.53
PassiveAggressiveClassifier 0.97 0.97 0.97 0.97 0.21
LinearDiscriminantAnalysis 0.97 0.96 0.96 0.97 0.19
CalibratedClassifierCV 0.97 0.96 0.96 0.97 0.24
AdaBoostClassifier 0.96 0.96 0.96 0.96 1.31
BaggingClassifier 0.95 0.95 0.95 0.95 0.63
RidgeClassifier 0.96 0.94 0.94 0.96 0.27
RidgeClassifierCV 0.96 0.94 0.94 0.96 0.18
QuadraticDiscriminantAnalysis 0.95 0.94 0.94 0.95 0.81
ExtraTreeClassifier 0.94 0.93 0.93 0.94 0.12
NuSVC 0.94 0.91 0.91 0.94 0.29
GaussianNB 0.93 0.91 0.91 0.93 0.17
BernoulliNB 0.92 0.90 0.90 0.92 0.31
NearestCentroid 0.92 0.89 0.89 0.92 0.24
DummyClassifier 0.64 0.50 0.50 0.50 0.27 
model_dictionary = clf.provide_models(X_train, X_test, y_train, y_test)

model_dictionary['LogisticRegression']

Pipeline(steps=[('preprocessor',
                 ColumnTransformer(transformers=[('numeric',
                                                  Pipeline(steps=[('imputer',
                                                                   SimpleImputer()),
                                                                  ('scaler',
                                                                   StandardScaler())]),
                                                  Int64Index([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,
            17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29],
           dtype='int64')),
                                                 ('categorical_low',
                                                  Pipeline(steps=[('imputer',
                                                                   SimpleImputer(fill_value='missing',
                                                                                 strategy='c...
                                                                   OneHotEncoder(handle_unknown='ignore',
                                                                                 sparse=False))]),
                                                  Int64Index([], dtype='int64')),
                                                 ('categorical_high',
                                                  Pipeline(steps=[('imputer',
                                                                   SimpleImputer(fill_value='missing',
                                                                                 strategy='constant')),
                                                                  ('encoding',
                                                                   OrdinalEncoder())]),
                                                  Int64Index([], dtype='int64'))])),
                ('classifier',
                 CustomClassifier(col_sample=0.9, n_hidden_features=10,
                                  obj=LogisticRegression(random_state=42)))])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
model_dictionary['LogisticRegression'].get_params()

{'memory': None,
 'steps': [('preprocessor',
   ColumnTransformer(transformers=[('numeric',
                                    Pipeline(steps=[('imputer', SimpleImputer()),
                                                    ('scaler', StandardScaler())]),
                                    Int64Index([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,
               17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29],
              dtype='int64')),
                                   ('categorical_low',
                                    Pipeline(steps=[('imputer',
                                                     SimpleImputer(fill_value='missing',
                                                                   strategy='constant')),
                                                    ('encoding',
                                                     OneHotEncoder(handle_unknown='ignore',
                                                                   sparse=False))]),
                                    Int64Index([], dtype='int64')),
                                   ('categorical_high',
                                    Pipeline(steps=[('imputer',
                                                     SimpleImputer(fill_value='missing',
                                                                   strategy='constant')),
                                                    ('encoding',
                                                     OrdinalEncoder())]),
                                    Int64Index([], dtype='int64'))])),
  ('classifier',
   CustomClassifier(col_sample=0.9, n_hidden_features=10,
                    obj=LogisticRegression(random_state=42)))],
 'verbose': False,
 'preprocessor': ColumnTransformer(transformers=[('numeric',
                                  Pipeline(steps=[('imputer', SimpleImputer()),
                                                  ('scaler', StandardScaler())]),
                                  Int64Index([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,
             17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29],
            dtype='int64')),
                                 ('categorical_low',
                                  Pipeline(steps=[('imputer',
                                                   SimpleImputer(fill_value='missing',
                                                                 strategy='constant')),
                                                  ('encoding',
                                                   OneHotEncoder(handle_unknown='ignore',
                                                                 sparse=False))]),
                                  Int64Index([], dtype='int64')),
                                 ('categorical_high',
                                  Pipeline(steps=[('imputer',
                                                   SimpleImputer(fill_value='missing',
                                                                 strategy='constant')),
                                                  ('encoding',
                                                   OrdinalEncoder())]),
                                  Int64Index([], dtype='int64'))]),
 'classifier': CustomClassifier(col_sample=0.9, n_hidden_features=10,
                  obj=LogisticRegression(random_state=42)),
 'preprocessor__n_jobs': None,
 'preprocessor__remainder': 'drop',
 'preprocessor__sparse_threshold': 0.3,
 'preprocessor__transformer_weights': None,
 'preprocessor__transformers': [('numeric',
   Pipeline(steps=[('imputer', SimpleImputer()), ('scaler', StandardScaler())]),
   Int64Index([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16,
               17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29],
              dtype='int64')),
  ('categorical_low',
   Pipeline(steps=[('imputer',
                    SimpleImputer(fill_value='missing', strategy='constant')),
                   ('encoding',
                    OneHotEncoder(handle_unknown='ignore', sparse=False))]),
   Int64Index([], dtype='int64')),
  ('categorical_high',
   Pipeline(steps=[('imputer',
                    SimpleImputer(fill_value='missing', strategy='constant')),
                   ('encoding', OrdinalEncoder())]),
   Int64Index([], dtype='int64'))],
 'preprocessor__verbose': False,
 'preprocessor__verbose_feature_names_out': True,
 'preprocessor__numeric': Pipeline(steps=[('imputer', SimpleImputer()), ('scaler', StandardScaler())]),
 'preprocessor__categorical_low': Pipeline(steps=[('imputer',
                  SimpleImputer(fill_value='missing', strategy='constant')),
                 ('encoding',
                  OneHotEncoder(handle_unknown='ignore', sparse=False))]),
 'preprocessor__categorical_high': Pipeline(steps=[('imputer',
                  SimpleImputer(fill_value='missing', strategy='constant')),
                 ('encoding', OrdinalEncoder())]),
 'preprocessor__numeric__memory': None,
 'preprocessor__numeric__steps': [('imputer', SimpleImputer()),
  ('scaler', StandardScaler())],
 'preprocessor__numeric__verbose': False,
 'preprocessor__numeric__imputer': SimpleImputer(),
 'preprocessor__numeric__scaler': StandardScaler(),
 'preprocessor__numeric__imputer__add_indicator': False,
 'preprocessor__numeric__imputer__copy': True,
 'preprocessor__numeric__imputer__fill_value': None,
 'preprocessor__numeric__imputer__keep_empty_features': False,
 'preprocessor__numeric__imputer__missing_values': nan,
 'preprocessor__numeric__imputer__strategy': 'mean',
 'preprocessor__numeric__imputer__verbose': 'deprecated',
 'preprocessor__numeric__scaler__copy': True,
 'preprocessor__numeric__scaler__with_mean': True,
 'preprocessor__numeric__scaler__with_std': True,
 'preprocessor__categorical_low__memory': None,
 'preprocessor__categorical_low__steps': [('imputer',
   SimpleImputer(fill_value='missing', strategy='constant')),
  ('encoding', OneHotEncoder(handle_unknown='ignore', sparse=False))],
 'preprocessor__categorical_low__verbose': False,
 'preprocessor__categorical_low__imputer': SimpleImputer(fill_value='missing', strategy='constant'),
 'preprocessor__categorical_low__encoding': OneHotEncoder(handle_unknown='ignore', sparse=False),
 'preprocessor__categorical_low__imputer__add_indicator': False,
 'preprocessor__categorical_low__imputer__copy': True,
 'preprocessor__categorical_low__imputer__fill_value': 'missing',
 'preprocessor__categorical_low__imputer__keep_empty_features': False,
 'preprocessor__categorical_low__imputer__missing_values': nan,
 'preprocessor__categorical_low__imputer__strategy': 'constant',
 'preprocessor__categorical_low__imputer__verbose': 'deprecated',
 'preprocessor__categorical_low__encoding__categories': 'auto',
 'preprocessor__categorical_low__encoding__drop': None,
 'preprocessor__categorical_low__encoding__dtype': numpy.float64,
 'preprocessor__categorical_low__encoding__handle_unknown': 'ignore',
 'preprocessor__categorical_low__encoding__max_categories': None,
 'preprocessor__categorical_low__encoding__min_frequency': None,
 'preprocessor__categorical_low__encoding__sparse': False,
 'preprocessor__categorical_low__encoding__sparse_output': True,
 'preprocessor__categorical_high__memory': None,
 'preprocessor__categorical_high__steps': [('imputer',
   SimpleImputer(fill_value='missing', strategy='constant')),
  ('encoding', OrdinalEncoder())],
 'preprocessor__categorical_high__verbose': False,
 'preprocessor__categorical_high__imputer': SimpleImputer(fill_value='missing', strategy='constant'),
 'preprocessor__categorical_high__encoding': OrdinalEncoder(),
 'preprocessor__categorical_high__imputer__add_indicator': False,
 'preprocessor__categorical_high__imputer__copy': True,
 'preprocessor__categorical_high__imputer__fill_value': 'missing',
 'preprocessor__categorical_high__imputer__keep_empty_features': False,
 'preprocessor__categorical_high__imputer__missing_values': nan,
 'preprocessor__categorical_high__imputer__strategy': 'constant',
 'preprocessor__categorical_high__imputer__verbose': 'deprecated',
 'preprocessor__categorical_high__encoding__categories': 'auto',
 'preprocessor__categorical_high__encoding__dtype': numpy.float64,
 'preprocessor__categorical_high__encoding__encoded_missing_value': nan,
 'preprocessor__categorical_high__encoding__handle_unknown': 'error',
 'preprocessor__categorical_high__encoding__unknown_value': None,
 'classifier__a': 0.01,
 'classifier__activation_name': 'relu',
 'classifier__backend': 'cpu',
 'classifier__bias': True,
 'classifier__cluster_encode': True,
 'classifier__col_sample': 0.9,
 'classifier__direct_link': True,
 'classifier__dropout': 0,
 'classifier__n_clusters': 2,
 'classifier__n_hidden_features': 10,
 'classifier__nodes_sim': 'sobol',
 'classifier__obj__C': 1.0,
 'classifier__obj__class_weight': None,
 'classifier__obj__dual': False,
 'classifier__obj__fit_intercept': True,
 'classifier__obj__intercept_scaling': 1,
 'classifier__obj__l1_ratio': None,
 'classifier__obj__max_iter': 100,
 'classifier__obj__multi_class': 'auto',
 'classifier__obj__n_jobs': None,
 'classifier__obj__penalty': 'l2',
 'classifier__obj__random_state': 42,
 'classifier__obj__solver': 'lbfgs',
 'classifier__obj__tol': 0.0001,
 'classifier__obj__verbose': 0,
 'classifier__obj__warm_start': False,
 'classifier__obj': LogisticRegression(random_state=42),
 'classifier__row_sample': 1,
 'classifier__seed': 123,
 'classifier__type_clust': 'kmeans',
 'classifier__type_scaling': ('std', 'std', 'std')}

3 - Regression

from sklearn.datasets import load_diabetes
from sklearn.model_selection import train_test_split

data = load_diabetes()
X = data.data
y= data.target
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = .2, random_state = 123)

regr = ns.LazyRegressor(verbose=0, ignore_warnings=True, custom_metric=None)
models, predictions = regr.fit(X_train, X_test, y_train, y_test)
model_dictionary = regr.provide_models(X_train, X_test, y_train, y_test)

100%|██████████| 40/40 [00:03<00:00, 12.38it/s]

display(models)
Adjusted R-Squared R-Squared RMSE Time Taken
Model
LassoLarsIC 0.53 0.59 51.11 0.03
SGDRegressor 0.53 0.58 51.24 0.03
HuberRegressor 0.53 0.58 51.26 0.05
Ridge 0.53 0.58 51.37 0.03
KernelRidge 0.53 0.58 51.37 0.03
RidgeCV 0.53 0.58 51.37 0.03
Lasso 0.52 0.58 51.52 0.03
LassoLars 0.52 0.58 51.52 0.03
LassoCV 0.52 0.58 51.58 0.12
LassoLarsCV 0.52 0.58 51.58 0.05
TransformedTargetRegressor 0.52 0.58 51.62 0.03
LinearRegression 0.52 0.58 51.62 0.03
OrthogonalMatchingPursuitCV 0.52 0.58 51.69 0.05
BayesianRidge 0.52 0.57 51.77 0.03
LinearSVR 0.51 0.57 52.04 0.02
ElasticNetCV 0.51 0.56 52.49 0.08
LarsCV 0.50 0.56 52.79 0.05
PassiveAggressiveRegressor 0.49 0.55 53.39 0.03
GradientBoostingRegressor 0.48 0.54 54.00 0.26
ElasticNet 0.46 0.52 54.92 0.03
BaggingRegressor 0.46 0.52 54.92 0.07
RandomForestRegressor 0.46 0.52 55.07 0.37
HistGradientBoostingRegressor 0.45 0.51 55.42 0.20
ExtraTreesRegressor 0.44 0.51 55.71 0.24
AdaBoostRegressor 0.44 0.51 55.75 0.14
MLPRegressor 0.43 0.50 56.38 0.45
TweedieRegressor 0.42 0.48 57.03 0.03
RANSACRegressor 0.42 0.48 57.14 0.16
KNeighborsRegressor 0.31 0.39 62.10 0.05
OrthogonalMatchingPursuit 0.31 0.38 62.27 0.04
GaussianProcessRegressor 0.19 0.28 67.13 0.05
ExtraTreeRegressor 0.15 0.24 69.09 0.03
SVR 0.12 0.22 69.98 0.04
NuSVR 0.12 0.22 70.14 0.04
DummyRegressor -0.13 -0.00 79.39 0.03
DecisionTreeRegressor -0.26 -0.11 83.75 0.03
Lars -1.95 -1.61 128.28 0.14 
model_dictionary["LassoLarsIC"]

Pipeline(steps=[('preprocessor',
                 ColumnTransformer(transformers=[('numeric',
                                                  Pipeline(steps=[('imputer',
                                                                   SimpleImputer()),
                                                                  ('scaler',
                                                                   StandardScaler())]),
                                                  Int64Index([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype='int64')),
                                                 ('categorical_low',
                                                  Pipeline(steps=[('imputer',
                                                                   SimpleImputer(fill_value='missing',
                                                                                 strategy='constant')),
                                                                  ('encoding',
                                                                   OneHotEncoder(handle_unknown='ignore',
                                                                                 sparse=False))]),
                                                  Int64Index([], dtype='int64')),
                                                 ('categorical_high',
                                                  Pipeline(steps=[('imputer',
                                                                   SimpleImputer(fill_value='missing',
                                                                                 strategy='constant')),
                                                                  ('encoding',
                                                                   OrdinalEncoder())]),
                                                  Int64Index([], dtype='int64'))])),
                ('regressor', CustomRegressor(obj=LassoLarsIC()))])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.
model_dictionary["LassoLarsIC"].get_params()

{'memory': None,
 'steps': [('preprocessor',
   ColumnTransformer(transformers=[('numeric',
                                    Pipeline(steps=[('imputer', SimpleImputer()),
                                                    ('scaler', StandardScaler())]),
                                    Int64Index([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype='int64')),
                                   ('categorical_low',
                                    Pipeline(steps=[('imputer',
                                                     SimpleImputer(fill_value='missing',
                                                                   strategy='constant')),
                                                    ('encoding',
                                                     OneHotEncoder(handle_unknown='ignore',
                                                                   sparse=False))]),
                                    Int64Index([], dtype='int64')),
                                   ('categorical_high',
                                    Pipeline(steps=[('imputer',
                                                     SimpleImputer(fill_value='missing',
                                                                   strategy='constant')),
                                                    ('encoding',
                                                     OrdinalEncoder())]),
                                    Int64Index([], dtype='int64'))])),
  ('regressor', CustomRegressor(obj=LassoLarsIC()))],
 'verbose': False,
 'preprocessor': ColumnTransformer(transformers=[('numeric',
                                  Pipeline(steps=[('imputer', SimpleImputer()),
                                                  ('scaler', StandardScaler())]),
                                  Int64Index([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype='int64')),
                                 ('categorical_low',
                                  Pipeline(steps=[('imputer',
                                                   SimpleImputer(fill_value='missing',
                                                                 strategy='constant')),
                                                  ('encoding',
                                                   OneHotEncoder(handle_unknown='ignore',
                                                                 sparse=False))]),
                                  Int64Index([], dtype='int64')),
                                 ('categorical_high',
                                  Pipeline(steps=[('imputer',
                                                   SimpleImputer(fill_value='missing',
                                                                 strategy='constant')),
                                                  ('encoding',
                                                   OrdinalEncoder())]),
                                  Int64Index([], dtype='int64'))]),
 'regressor': CustomRegressor(obj=LassoLarsIC()),
 'preprocessor__n_jobs': None,
 'preprocessor__remainder': 'drop',
 'preprocessor__sparse_threshold': 0.3,
 'preprocessor__transformer_weights': None,
 'preprocessor__transformers': [('numeric',
   Pipeline(steps=[('imputer', SimpleImputer()), ('scaler', StandardScaler())]),
   Int64Index([0, 1, 2, 3, 4, 5, 6, 7, 8, 9], dtype='int64')),
  ('categorical_low',
   Pipeline(steps=[('imputer',
                    SimpleImputer(fill_value='missing', strategy='constant')),
                   ('encoding',
                    OneHotEncoder(handle_unknown='ignore', sparse=False))]),
   Int64Index([], dtype='int64')),
  ('categorical_high',
   Pipeline(steps=[('imputer',
                    SimpleImputer(fill_value='missing', strategy='constant')),
                   ('encoding', OrdinalEncoder())]),
   Int64Index([], dtype='int64'))],
 'preprocessor__verbose': False,
 'preprocessor__verbose_feature_names_out': True,
 'preprocessor__numeric': Pipeline(steps=[('imputer', SimpleImputer()), ('scaler', StandardScaler())]),
 'preprocessor__categorical_low': Pipeline(steps=[('imputer',
                  SimpleImputer(fill_value='missing', strategy='constant')),
                 ('encoding',
                  OneHotEncoder(handle_unknown='ignore', sparse=False))]),
 'preprocessor__categorical_high': Pipeline(steps=[('imputer',
                  SimpleImputer(fill_value='missing', strategy='constant')),
                 ('encoding', OrdinalEncoder())]),
 'preprocessor__numeric__memory': None,
 'preprocessor__numeric__steps': [('imputer', SimpleImputer()),
  ('scaler', StandardScaler())],
 'preprocessor__numeric__verbose': False,
 'preprocessor__numeric__imputer': SimpleImputer(),
 'preprocessor__numeric__scaler': StandardScaler(),
 'preprocessor__numeric__imputer__add_indicator': False,
 'preprocessor__numeric__imputer__copy': True,
 'preprocessor__numeric__imputer__fill_value': None,
 'preprocessor__numeric__imputer__keep_empty_features': False,
 'preprocessor__numeric__imputer__missing_values': nan,
 'preprocessor__numeric__imputer__strategy': 'mean',
 'preprocessor__numeric__imputer__verbose': 'deprecated',
 'preprocessor__numeric__scaler__copy': True,
 'preprocessor__numeric__scaler__with_mean': True,
 'preprocessor__numeric__scaler__with_std': True,
 'preprocessor__categorical_low__memory': None,
 'preprocessor__categorical_low__steps': [('imputer',
   SimpleImputer(fill_value='missing', strategy='constant')),
  ('encoding', OneHotEncoder(handle_unknown='ignore', sparse=False))],
 'preprocessor__categorical_low__verbose': False,
 'preprocessor__categorical_low__imputer': SimpleImputer(fill_value='missing', strategy='constant'),
 'preprocessor__categorical_low__encoding': OneHotEncoder(handle_unknown='ignore', sparse=False),
 'preprocessor__categorical_low__imputer__add_indicator': False,
 'preprocessor__categorical_low__imputer__copy': True,
 'preprocessor__categorical_low__imputer__fill_value': 'missing',
 'preprocessor__categorical_low__imputer__keep_empty_features': False,
 'preprocessor__categorical_low__imputer__missing_values': nan,
 'preprocessor__categorical_low__imputer__strategy': 'constant',
 'preprocessor__categorical_low__imputer__verbose': 'deprecated',
 'preprocessor__categorical_low__encoding__categories': 'auto',
 'preprocessor__categorical_low__encoding__drop': None,
 'preprocessor__categorical_low__encoding__dtype': numpy.float64,
 'preprocessor__categorical_low__encoding__handle_unknown': 'ignore',
 'preprocessor__categorical_low__encoding__max_categories': None,
 'preprocessor__categorical_low__encoding__min_frequency': None,
 'preprocessor__categorical_low__encoding__sparse': False,
 'preprocessor__categorical_low__encoding__sparse_output': True,
 'preprocessor__categorical_high__memory': None,
 'preprocessor__categorical_high__steps': [('imputer',
   SimpleImputer(fill_value='missing', strategy='constant')),
  ('encoding', OrdinalEncoder())],
 'preprocessor__categorical_high__verbose': False,
 'preprocessor__categorical_high__imputer': SimpleImputer(fill_value='missing', strategy='constant'),
 'preprocessor__categorical_high__encoding': OrdinalEncoder(),
 'preprocessor__categorical_high__imputer__add_indicator': False,
 'preprocessor__categorical_high__imputer__copy': True,
 'preprocessor__categorical_high__imputer__fill_value': 'missing',
 'preprocessor__categorical_high__imputer__keep_empty_features': False,
 'preprocessor__categorical_high__imputer__missing_values': nan,
 'preprocessor__categorical_high__imputer__strategy': 'constant',
 'preprocessor__categorical_high__imputer__verbose': 'deprecated',
 'preprocessor__categorical_high__encoding__categories': 'auto',
 'preprocessor__categorical_high__encoding__dtype': numpy.float64,
 'preprocessor__categorical_high__encoding__encoded_missing_value': nan,
 'preprocessor__categorical_high__encoding__handle_unknown': 'error',
 'preprocessor__categorical_high__encoding__unknown_value': None,
 'regressor__a': 0.01,
 'regressor__activation_name': 'relu',
 'regressor__backend': 'cpu',
 'regressor__bias': True,
 'regressor__cluster_encode': True,
 'regressor__col_sample': 1,
 'regressor__direct_link': True,
 'regressor__dropout': 0,
 'regressor__n_clusters': 2,
 'regressor__n_hidden_features': 5,
 'regressor__nodes_sim': 'sobol',
 'regressor__obj__copy_X': True,
 'regressor__obj__criterion': 'aic',
 'regressor__obj__eps': 2.220446049250313e-16,
 'regressor__obj__fit_intercept': True,
 'regressor__obj__max_iter': 500,
 'regressor__obj__noise_variance': None,
 'regressor__obj__normalize': 'deprecated',
 'regressor__obj__positive': False,
 'regressor__obj__precompute': 'auto',
 'regressor__obj__verbose': False,
 'regressor__obj': LassoLarsIC(),
 'regressor__row_sample': 1,
 'regressor__seed': 123,
 'regressor__type_clust': 'kmeans',
 'regressor__type_scaling': ('std', 'std', 'std')}

Comments powered by Talkyard.