Hi, I just read the "Cluster Contrast" paper on arxiv and found it a creative work. However, I am still confused on the ablation studies of Table 4, how do you fix the cluster size to a constant number since it can not be pre-defined in DBSCAN clustering method?
Hi, I just read the "Cluster Contrast" paper on arxiv and found it a creative work. However, I am still confused on the ablation studies of Table 4, how do you fix the cluster size to a constant number since it can not be pre-defined in DBSCAN clustering method?