GridSearchCV using Pipeline - Plot the ROC curve of every fold - matplotlib

Question

I would like to plot the ROC curve of every fold in a RepeatedStratifiedKFold. I am using the Pipeline class to fit my models like this:
def cross_validation(df, clf, param_grid, n_splits=5, n_repeats=50):
# Get data
features = df_features.loc[:, ~df_features.columns.isin(['PATNO', 'output', 'Imaging Protocol'])].columns
X = df[features]
y = df["output"]
# Define CV
cv = RepeatedStratifiedKFold(n_splits=n_splits, n_repeats=n_repeats, random_state=42)
# Define pipeline
pipeline = Pipeline([
('scale', StandardScaler()),
('train', clf)
])
# Define grid search
grid = GridSearchCV(pipeline, param_grid, cv=cv, scoring='roc_auc', return_train_score=True, n_jobs=-1)
# Train
grid.fit(X, y)
return grid
I am plotting my ROC curve this way:
def display_roc_cv_curves(title):
plt.plot([0, 1], [0, 1], "k--", label="chance level (AUC = 0.5)")
plt.xlim([0.0, 1.0])
plt.ylim([0.0, 1.05])
plt.xlabel('1-Specificity(False Positive Rate)')
plt.ylabel('Sensitivity(True Positive Rate)')
plt.title('Receiver Operating Characteristic')
plt.legend(loc="lower right", prop={'size': 10})
plt.title(title)
plt.savefig(f"outputs/performance/{COHORT_TYPE}/{FEATURE_TYPE}_cv.png")
plt.show()
plt.figure(figsize=(10,8))
for model in modelPerformance:
X_test = test[features]
y_test = test["output"]
bestModel = model["trainedModel"]
modelName = type(bestModel.best_estimator_["train"]).__name__
y_pred_proba = bestModel.decision_function(X_test) if modelName=="SVM" else bestModel.predict_proba(X_test)[:, 1]
fpr, tpr, _ = metrics.roc_curve(y_test, y_pred_proba)
auc = metrics.roc_auc_score(y_test, y_pred_proba)
plt.plot(fpr, tpr, label=f'{modelName} AUC = {round(auc, 3)}')
display_roc_cv_curves(f"CV ROC of {COHORT_TYPE} with {FEATURE_TYPE}")
This works well and shows the ROC curve of every model on the test set. However, I'd like to create an additional plot where I plot the ROC curve of every fold in shaded gray in the background along with the average ROC of the validation set. This is an example representing what I want. I saw that this can be done if we iterate through the CV loop and manually add plotting. In this case, I am using a GridSearch so I am not sure how to accomplish this.