Evaluating Identification 1:N with multiple images per Identity in gallery

Hello Everyone,

If in case there are multiple images per person in the gallery. Typically say at different poses , during evalution in 1:N what strategies we can use to get the final score/scores.

Probe = 10 images (Each person 1 image, Total 10 person,IDs = 21 to 29 and ID = 4)
Gallery = 5 folders (Each folder contains 10 images , Total 5 person & 50 images , IDs = 0 to 4)
Note: Each ID belongs to one unique person. In the above dataset probe image of ID 4 is same identity/person as in Gallery of ID 4

Lets say probe image of ID 4 is macthed with the gallery,i will get 1x50 scores. Now how to select the Top1 result. Within Gallery of ID 4,as there are multiple images per person, what startegy to follow to arrive at top1 result. Does the current evaluation_1N function in [here](https://github.com/deepinsight/insightface/blob/master/recognition/_evaluation_/ijb/ijb_evals.py#:~:text=def%20evaluation_1N(query_feats%2C%20gallery_feats%2C%20query_ids%2C%20reg_ids%2C%20fars%3D%5B0.01%2C%200.1%5D)%3A) supports this type of evaluations as well ?