Solving a binary classification problem. How will precision and recall metrics change if you remove 10 class 0 objects from the target?

If 10 objects of class 0 (negative class) are removed from the binary classification target, the class distribution will change, which will affect the precision and recall metrics as follows:

• Precision — this is the ratio of true positives (TP) to the sum of TP and false positives (FP): TP / (TP + FP). Removing class 0 objects from the target does not directly affect TP or FP, as these metrics depend on the model's predictions. If the predictions do not change, the precision will remain approximately the same.

• Recall — this is the ratio of TP to the sum of TP and false negatives (FN): TP / (TP + FN). Removing class 0 objects does not affect the number of TP or FN for class 1, so the recall for class 1 will remain approximately the same.

However, if we consider metrics for class 0 (negative class), removing class 0 objects will decrease the number of true negatives (TN) and potentially affect FP and FN, which may change the precision and recall for this class.

Overall, removing class 0 objects from the target will change the class distribution, which may influence the metrics, especially if they are averaged over classes (e.g., macro-averaged). But for metrics calculated only for the positive class, the changes will be minimal if predictions do not change.