search

A4Bio / RDesign

The official implementation of the ICLR'24 paper RDesign: Hierarchical Data-efficient Representation Learning for Tertiary Structure-based RNA Design.
36 stars 0 forks source link

Question about the metrics #3

Open PAIplus opened 6 months ago

PAIplus commented 6 months ago

https://github.com/A4Bio/RDesign/blob/b61988824d9e69cde316e0c593c446711f0ce36f/methods/rdesign.py#L60 We found that "np.median" is used in the metrics. Is it the result reported in the paper? Or we should change it into np.mean ?

chengtan9907 commented 6 months ago

Yes. We use np.median for each experiment, and repeat three times with different randoms. Then, we present the average and standard deviation of these three iterations as shown in the paper.