# ROC AUC for multiclass problem

Just some quick questions to clarify my doubt please.

I know that one can get precision/recall for each class in a multiclass problem, e.g. in this classification report:

Classification report
precision    recall  f1-score   support

0       0.65      0.62      0.63     14601
1       0.07      0.27      0.11      1398
2       0.43      0.06      0.10      8317
3       0.58      0.70      0.64     20301
4       0.00      0.00      0.00       904

accuracy                           0.53     45521
macro avg       0.34      0.33      0.30     45521
weighted avg       0.55      0.53      0.51     45521


Accuracy is not computed as per class, since it's a global metric.

1. Can one also computes ROC AUC for each class.
2. What the difference between macro/weighted avg precision/recall and the value for each class. Is that representing a global measure of precision/recall?