Skip to content

naive bayes

Module: naive_bayes.py

This module implements Naive Bayes classification algorithms, including: - Gaussian Naive Bayes - Multinomial Naive Bayes - Bernoulli Naive Bayes

It provides functionality to
  • Train Naive Bayes classifiers on a dataset
  • Evaluate model performance using classification metrics
  • Visualize results with a confusion matrix
  • Optimize hyperparameters using grid search

Classes:

Name Description
NaiveBayes

Implements Naive Bayes classification using scikit-learn.
Dependencies
  • numpy
  • sklearn.naive_bayes.GaussianNB
  • sklearn.naive_bayes.MultinomialNB
  • sklearn.naive_bayes.BernoulliNB
  • sklearn.metrics (classification metrics)
  • base.py (Classification)
Key Features
  • Support for Gaussian, Multinomial, and Bernoulli Naive Bayes
  • Grid search for hyperparameter tuning
  • Automatic data preparation and evaluation
Authors
  • Protyush P. Chowdhury (protyushc@iisc.ac.in)
Version Info
  • 28/Dec/2024: Initial version

NaiveBayes

Bases: Classification

Implements Naive Bayes classification for Gaussian, Multinomial, and Bernoulli distributions.

Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
class NaiveBayes(Classification):
    """
    Implements Naive Bayes classification for Gaussian, Multinomial, and Bernoulli distributions.
    """

    def __init__(self, model_type: str = "gaussian", random_state: int = 42) -> None:
        """
        Initialize the NaiveBayes classifier.

        Args:
            model_type (str): Type of Naive Bayes classifier. Options are:
                              "gaussian", "multinomial", "bernoulli".
            random_state (int): Seed for reproducibility where applicable.
        """
        super().__init__(model_type=model_type, random_state=random_state)

        if model_type == "gaussian":
            self.model = GaussianNB()
        elif model_type == "multinomial":
            self.model = MultinomialNB()
        elif model_type == "bernoulli":
            self.model = BernoulliNB()
        else:
            raise ValueError(
                "Invalid model_type. Choose 'gaussian', 'multinomial', or 'bernoulli'."
            )

    def fit(self, X_train: np.ndarray, y_train: np.ndarray) -> None:
        """
        Train the Naive Bayes model.

        Args:
            X_train (np.ndarray): Training data features.
            y_train (np.ndarray): Training data labels.
        """
        self.model.fit(X_train, y_train)

    def evaluate(self, X_test: np.ndarray, y_test: np.ndarray) -> Dict[str, Any]:
        """
        Evaluate the model on test data.

        Args:
            X_test (np.ndarray): Test data features.
            y_test (np.ndarray): Test data labels.

        Returns:
            Dict[str, Any]: Dictionary containing evaluation metrics (accuracy, precision, recall, F1-score).
        """
        y_pred = self.model.predict(X_test)
        report = classification_report(y_test, y_pred, output_dict=True)
        return {
            "accuracy": report["accuracy"],
            "precision": report["weighted avg"]["precision"],
            "recall": report["weighted avg"]["recall"],
            "f1_score": report["weighted avg"]["f1-score"],
        }

    def plot(self, X_test: np.ndarray, y_test: np.ndarray) -> None:
        """
        Plot the confusion matrix for the test data.

        Args:
            X_test (np.ndarray): Test data features.
            y_test (np.ndarray): Test data labels.
        """
        y_pred = self.model.predict(X_test)
        cm = confusion_matrix(y_test, y_pred)
        plt.figure(figsize=(8, 6))
        sns.heatmap(
            cm,
            annot=True,
            fmt="d",
            cmap="Blues",
            xticklabels=np.unique(y_test),
            yticklabels=np.unique(y_test),
        )
        plt.xlabel("Predicted")
        plt.ylabel("Actual")
        plt.title(f"Confusion Matrix - {self.model_type.capitalize()} Naive Bayes")
        plt.show()

    def grid_search(
        self,
        X_train: np.ndarray,
        y_train: np.ndarray,
        param_grid: Dict[str, Any],
        cv: int = 5,
    ) -> None:
        """
        Perform hyperparameter tuning using grid search.

        Args:
            X_train (np.ndarray): Training data features.
            y_train (np.ndarray): Training data labels.
            param_grid (Dict[str, Any]): Dictionary of hyperparameters to search.
            cv (int): Number of cross-validation folds. Default is 5.
        """
        grid = GridSearchCV(
            estimator=self.model, param_grid=param_grid, scoring="accuracy", cv=cv
        )
        grid.fit(X_train, y_train)
        self.model = grid.best_estimator_
        print(f"Best Parameters: {grid.best_params_}")
        print(f"Best Cross-Validated Accuracy: {grid.best_score_}")

    def run(
        self,
        data: np.ndarray,
        labels: np.ndarray,
        test_size: float = 0.2,
        param_grid: Dict[str, Any] = None,
        cv: int = 5,
    ) -> Dict[str, Any]:
        """
        Execute the full classification pipeline with optional grid search.

        Args:
            data (np.ndarray): Input features.
            labels (np.ndarray): Input labels.
            test_size (float): Proportion of data to use for testing. Defaults to 0.2.
            param_grid (Dict[str, Any], optional): Dictionary of hyperparameters to search for grid search. If None, grid search will not be performed.
            cv (int): Number of cross-validation folds for grid search. Default is 5.

        Returns:
            Dict[str, Any]: Performance metrics.
        """
        # Split the data into training and test sets
        X_train, X_test, y_train, y_test = train_test_split(
            data, labels, test_size=test_size, random_state=self.random_state
        )

        # Perform grid search if param_grid is provided
        if param_grid:
            self.grid_search(X_train, y_train, param_grid, cv=cv)

        # Train the model with the (possibly tuned) parameters
        self.fit(X_train, y_train)

        # Evaluate the model
        metrics = self.evaluate(X_test, y_test)

        # Plot the confusion matrix
        self.plot(X_test, y_test)

        return metrics

    # Implementing the abstract method get_model_params
    def get_model_params(self) -> Dict[str, Any]:
        """
        Return the parameters of the model.

        Returns:
            Dict[str, Any]: Dictionary containing model parameters.
        """
        return {
            "alpha": self.model.alpha,
            "fit_prior": self.model.fit_prior,
            "class_prior": self.model.class_prior,
        }

    # Method to save the model to a file
    def save_model(self, file_path: str) -> None:
        """
        Save the trained model to a file.

        Args:
            file_path (str): Path where the model will be saved.
        """
        joblib.dump(self.model, file_path)
        print(f"Model saved to {file_path}")

__init__(model_type='gaussian', random_state=42)

Initialize the NaiveBayes classifier.

Parameters:

Name Type Description Default
model_type str

Type of Naive Bayes classifier. Options are: "gaussian", "multinomial", "bernoulli".

 'gaussian'
random_state int

Seed for reproducibility where applicable.

 42
Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
def __init__(self, model_type: str = "gaussian", random_state: int = 42) -> None:
    """
    Initialize the NaiveBayes classifier.

    Args:
        model_type (str): Type of Naive Bayes classifier. Options are:
                          "gaussian", "multinomial", "bernoulli".
        random_state (int): Seed for reproducibility where applicable.
    """
    super().__init__(model_type=model_type, random_state=random_state)

    if model_type == "gaussian":
        self.model = GaussianNB()
    elif model_type == "multinomial":
        self.model = MultinomialNB()
    elif model_type == "bernoulli":
        self.model = BernoulliNB()
    else:
        raise ValueError(
            "Invalid model_type. Choose 'gaussian', 'multinomial', or 'bernoulli'."
        )

evaluate(X_test, y_test)

Evaluate the model on test data.

Parameters:

Name Type Description Default
X_test ndarray

Test data features.

 required
y_test ndarray

Test data labels.

 required

Returns:

Type Description
Dict[str, Any]

Dict[str, Any]: Dictionary containing evaluation metrics (accuracy, precision, recall, F1-score).
Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
def evaluate(self, X_test: np.ndarray, y_test: np.ndarray) -> Dict[str, Any]:
    """
    Evaluate the model on test data.

    Args:
        X_test (np.ndarray): Test data features.
        y_test (np.ndarray): Test data labels.

    Returns:
        Dict[str, Any]: Dictionary containing evaluation metrics (accuracy, precision, recall, F1-score).
    """
    y_pred = self.model.predict(X_test)
    report = classification_report(y_test, y_pred, output_dict=True)
    return {
        "accuracy": report["accuracy"],
        "precision": report["weighted avg"]["precision"],
        "recall": report["weighted avg"]["recall"],
        "f1_score": report["weighted avg"]["f1-score"],
    }

fit(X_train, y_train)

Train the Naive Bayes model.

Parameters:

Name Type Description Default
X_train ndarray

Training data features.

 required
y_train ndarray

Training data labels.

 required
Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
104
105
106
107
108
109
110
111
112
def fit(self, X_train: np.ndarray, y_train: np.ndarray) -> None:
    """
    Train the Naive Bayes model.

    Args:
        X_train (np.ndarray): Training data features.
        y_train (np.ndarray): Training data labels.
    """
    self.model.fit(X_train, y_train)

get_model_params()

Return the parameters of the model.

Returns:

Type Description
Dict[str, Any]

Dict[str, Any]: Dictionary containing model parameters.
Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
224
225
226
227
228
229
230
231
232
233
234
235
def get_model_params(self) -> Dict[str, Any]:
    """
    Return the parameters of the model.

    Returns:
        Dict[str, Any]: Dictionary containing model parameters.
    """
    return {
        "alpha": self.model.alpha,
        "fit_prior": self.model.fit_prior,
        "class_prior": self.model.class_prior,
    }

Perform hyperparameter tuning using grid search.

Parameters:

Name Type Description Default
X_train ndarray

Training data features.

 required
y_train ndarray

Training data labels.

 required
param_grid Dict[str, Any]

Dictionary of hyperparameters to search.

 required
cv int

Number of cross-validation folds. Default is 5.

 5
Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
def grid_search(
    self,
    X_train: np.ndarray,
    y_train: np.ndarray,
    param_grid: Dict[str, Any],
    cv: int = 5,
) -> None:
    """
    Perform hyperparameter tuning using grid search.

    Args:
        X_train (np.ndarray): Training data features.
        y_train (np.ndarray): Training data labels.
        param_grid (Dict[str, Any]): Dictionary of hyperparameters to search.
        cv (int): Number of cross-validation folds. Default is 5.
    """
    grid = GridSearchCV(
        estimator=self.model, param_grid=param_grid, scoring="accuracy", cv=cv
    )
    grid.fit(X_train, y_train)
    self.model = grid.best_estimator_
    print(f"Best Parameters: {grid.best_params_}")
    print(f"Best Cross-Validated Accuracy: {grid.best_score_}")

plot(X_test, y_test)

Plot the confusion matrix for the test data.

Parameters:

Name Type Description Default
X_test ndarray

Test data features.

 required
y_test ndarray

Test data labels.

 required
Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
def plot(self, X_test: np.ndarray, y_test: np.ndarray) -> None:
    """
    Plot the confusion matrix for the test data.

    Args:
        X_test (np.ndarray): Test data features.
        y_test (np.ndarray): Test data labels.
    """
    y_pred = self.model.predict(X_test)
    cm = confusion_matrix(y_test, y_pred)
    plt.figure(figsize=(8, 6))
    sns.heatmap(
        cm,
        annot=True,
        fmt="d",
        cmap="Blues",
        xticklabels=np.unique(y_test),
        yticklabels=np.unique(y_test),
    )
    plt.xlabel("Predicted")
    plt.ylabel("Actual")
    plt.title(f"Confusion Matrix - {self.model_type.capitalize()} Naive Bayes")
    plt.show()

run(data, labels, test_size=0.2, param_grid=None, cv=5)

Execute the full classification pipeline with optional grid search.

Parameters:

Name Type Description Default
data ndarray

Input features.

 required
labels ndarray

Input labels.

 required
test_size float

Proportion of data to use for testing. Defaults to 0.2.

 0.2
param_grid Dict[str, Any]

Dictionary of hyperparameters to search for grid search. If None, grid search will not be performed.

 None
cv int

Number of cross-validation folds for grid search. Default is 5.

 5

Returns:

Type Description
Dict[str, Any]

Dict[str, Any]: Performance metrics.
Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
def run(
    self,
    data: np.ndarray,
    labels: np.ndarray,
    test_size: float = 0.2,
    param_grid: Dict[str, Any] = None,
    cv: int = 5,
) -> Dict[str, Any]:
    """
    Execute the full classification pipeline with optional grid search.

    Args:
        data (np.ndarray): Input features.
        labels (np.ndarray): Input labels.
        test_size (float): Proportion of data to use for testing. Defaults to 0.2.
        param_grid (Dict[str, Any], optional): Dictionary of hyperparameters to search for grid search. If None, grid search will not be performed.
        cv (int): Number of cross-validation folds for grid search. Default is 5.

    Returns:
        Dict[str, Any]: Performance metrics.
    """
    # Split the data into training and test sets
    X_train, X_test, y_train, y_test = train_test_split(
        data, labels, test_size=test_size, random_state=self.random_state
    )

    # Perform grid search if param_grid is provided
    if param_grid:
        self.grid_search(X_train, y_train, param_grid, cv=cv)

    # Train the model with the (possibly tuned) parameters
    self.fit(X_train, y_train)

    # Evaluate the model
    metrics = self.evaluate(X_test, y_test)

    # Plot the confusion matrix
    self.plot(X_test, y_test)

    return metrics

save_model(file_path)

Save the trained model to a file.

Parameters:

Name Type Description Default
file_path str

Path where the model will be saved.

 required
Source code in scirex/core/ml/supervised/classification/naive_bayes.py 
238
239
240
241
242
243
244
245
246
def save_model(self, file_path: str) -> None:
    """
    Save the trained model to a file.

    Args:
        file_path (str): Path where the model will be saved.
    """
    joblib.dump(self.model, file_path)
    print(f"Model saved to {file_path}")