search

kakao / n2

TOROS N2 - lightweight approximate Nearest Neighbor library which runs fast even with large datasets
Apache License 2.0
569 stars 70 forks source link

With the same accuracy, the number of comparisons with the paper is very different. #31

Closed MartinNum closed 4 years ago

MartinNum commented 4 years ago

In the case of 90% accuracy, the paper only needs to compare hundreds of times, why is your code compared thousands of times?

gony-noreply commented 4 years ago

Could you tell me the dataset and n2 parameter you used? and I wonder how you measured the number of comparisons.

MartinNum commented 4 years ago

Could you tell me the dataset and n2 parameter you used? and I wonder how you measured the number of comparisons.

It’s my problem. The random dataset I used resulted in a high number of comparisons. After using the dataset SIFT1M, the comparison times became normal.

gony-noreply commented 4 years ago

Sorry for the late feedback, we will close the issue.