Update example notebook

examples/binary_classifier.md

@@ -27,7 +27,7 @@ from sklearn.ensemble import RandomForestClassifier
 from sklearn.linear_model import LogisticRegression
 from sklearn.model_selection import cross_val_predict
 
-from ml_inspector import plot_precision_recall_curves, plot_roc_curves, plot_gain_curves
+from ml_inspector import *
 ```
 
 ## Load classification dataset
@@ -45,7 +45,8 @@ y = pd.Series(dataset["target"])
 ```
 
 ```python
-class_names = dataset["target_names"]
+class_names = {i: str(c) for i, c in enumerate(dataset["target_names"])}
+class_names
 ```
 
 ## Train binary classification model
@@ -60,6 +61,13 @@ rf.fit(X, y)
 
 ## Make predictions
 
+```python
+y_pred = {
+    "Training": rf.predict(X), 
+    "Cross-Validation": cross_val_predict(rf, X, y, cv=5)
+}
+```
+
 ```python
 y_prob = {
     "Training": rf.predict_proba(X), 
@@ -84,3 +92,37 @@ plot_precision_recall_curves(y, y_prob, class_names, decision_threshold=0.5)
 ```python
 plot_gain_curves(y, y_prob, class_names, decision_threshold=0.5)
 ```
+
+## Confusion matrix
+
+```python
+plot_confusion_matrix(y, y_pred["Cross-Validation"], class_names)
+```
+
+## Probability distributions
+
+```python
+plot_classification_predictions(y, y_prob["Cross-Validation"], class_names, decision_threshold=0.5, points="all")
+```
+
+## Class calibration curves
+
+```python
+plot_calibration_curves(y, y_prob["Cross-Validation"], class_names, n_bins=10)
+```
+
+## Learning curve
+
+```python
+plot_learning_curves(rf, X, y, scoring="roc_auc")
+```
+
+## Partial dependence
+
+```python
+plot_partial_dependence(rf, X, "worst concavity")
+```
+
+```python
+
+```