bayes_models package¶
Submodules¶
bayes_models.model module¶
-
class
bayes_models.model.BGLClassifier(numeric_cols, target_col, cat_cols)¶ Bases:
bayes_models.model._AssembleClassificationModelCode,bayes_models.model._SSModelBaseThis class is responsible for performing Bayesian-Gaussian Logistic Regression, which treats categorical variables correctly.
-
fit(df)¶ This method is responsible for fitting the model onto the data
- Parameters
df (pandas.DataFrame) – Our dataset with our features and predictive columns
- Returns
self
-
predict(test_df)¶ This predicts the binary values of our test dataframe
- Parameters
test_df (pandas.DataFrame) – The dataset we will predict on.
- Returns
res
-
predict_proba(test_df)¶ This predicts the raw probabilities of our test dataframe
- Parameters
test_df (pandas.DataFrame) – The dataset we will predict on.
- Returns
res
-
score(x_val, y_val)¶ This returns the accuracy of the predictions
- Parameters
x_val (pandas.DataFrame) – The validation dataset
y_val (numpy.array or pandas.DataFrame) – The validation actual values
- Returns
-
-
class
bayes_models.model.BGLRegressor(numeric_cols, target_col, cat_cols)¶ Bases:
bayes_models.model._AssembleRegressionModelCode,bayes_models.model._SSModelBaseThis class is responsible for performing Bayesian-Gaussian Linear Regression, which treats categorical variables correctly.
-
fit(df)¶ This method is responsible for fitting the model onto the data
- Parameters
df (pandas.DataFrame) – Our dataset with our features and predictive columns
- Returns
self
-
predict(test_df)¶ This predicts the raw values of our test dataframe
- Parameters
test_df (pandas.DataFrame) – The dataset we will predict on.
- Returns
res
-
score(x_val, y_val)¶ The score returned is the R2 Score between the predicted and the actual.
- Parameters
x_val (pandas.DataFrame) – The validation dataset
y_val (numpy.array or pandas.DataFrame) – The validation actual values
- Returns
R2 Score of Predictions
-
-
class
bayes_models.model.BGRFClassifier(train_cols, target_col, n_estimators, max_depth)¶ Bases:
bayes_models.model._AssembleClassificationModelCode,bayes_models.model._SSModelBaseThis class is responsible for performing Bayesian-Gaussian Logistic Regression on the predictions of the Decision Trees from the fitted sklearn.ensemble.RandomForestClassifier model. This helps us optimally derive the best weighting based on the predictions of the individual Decision Trees.
-
fit(df)¶ First, we will fit the RandomForest on our dataset. Then we will use those predictions as features for our Bayesian Model
- Parameters
df (pandas.DataFrame) – Our dataset with our features and predictive columns
- Returns
self
-
predict(test_df)¶ This predicts the binary values of our test dataframe
- Parameters
test_df (pandas.DataFrame) – The dataset we will predict on.
- Returns
res
-
predict_proba(test_df)¶ This predicts the raw values of our test dataframe
- Parameters
test_df (pandas.DataFrame) – The dataset we will predict on.
- Returns
res
-
score(x_val, y_val)¶ This returns the accuracy of the predictions
- Parameters
x_val (pandas.DataFrame) – The validation dataset
y_val (numpy.array or pandas.DataFrame) – The validation actual values
- Returns
-
-
class
bayes_models.model.BGRFRegressor(train_cols, target_col, n_estimators, max_depth)¶ Bases:
bayes_models.model._AssembleRegressionModelCode,bayes_models.model._SSModelBaseThis class is responsible for performing Bayesian-Gaussian Linear Regression on the predictions of the Decision Trees from the fitted sklearn.ensemble.RandomForestRegressor model. This helps us optimally derive the best weighting based on the predictions of the individual Decision Trees.
-
fit(df)¶ First, we will fit the RandomForest on our dataset. Then we will use those predictions as features for our Bayesian Model
- Parameters
df (pandas.DataFrame) – Our dataset with our features and predictive columns
- Returns
self
-
predict(test_df)¶ This predicts the raw values of our test dataframe
- Parameters
test_df (pandas.DataFrame) – The dataset we will predict on.
- Returns
res
-
score(x_val, y_val)¶ The score returned is the R2 Score between the predicted and the actual.
- Parameters
x_val (pandas.DataFrame) – The validation dataset
y_val (numpy.array or pandas.DataFrame) – The validation actual values
- Returns
-