aggregation

a cluster of aggregation algorithm

KNN

• train set size = 10000
• test set size = 1000
• k = 10
• final accuracy = 0.755
• cost time = 191s

KMeans

• find k center:
  • train set size = 10000
  • test set size = 1000
  • k = 10 iter = 30 final accuracy = 0.563 cost time = 82s
  • k = 30 iter = 30 final accuracy = 0.713 cost time = 222s