MinMaxScaler »ç¿ë¹ý¿¡ ´ëÇؼ ¾Ë¾Æ º¸ÀÚ. MinMaxScaler´Â °ªÀ» 0°ú 1»çÀÌÀÇ °ªÀ¸·Î Á¤±ÔÈ ÇÑ´Ù. °ø½ÄÀº ´ÙÀ½°ú °°´Ù. y = (x - minx) / (maxx -minx) MinMaxScaler °ªÀº 1Â÷¿ø ¹è¿À» ¾È¹Þ±â ¶§¹®¿¡ 2Â÷¿øÀ¸·Î ¹è¿À» ¸¸µé¾ú´Ù. from sklearn.preprocessing import MinMaxScaler
data = [[0, 1000], [10, 1], [20, 1], [30, 1], [40, 1], [50, 1]] scaler = MinMaxScaler() res = scaler.fit(data) print('----------------------') print('min=', scaler.data_min_) print('max=', scaler.data_max_) print('------ transform --------') res = scaler.transform(data) print(res) print('res[4]= ', res[4]) print('res[5]= ', res[5][0]) °á°ú) ---------------------- min= [0. 1.] max= [ 50. 1000.] ------ transform -------- [[0. 1. ] [0.2 0. ] [0.4 0. ] [0.6 0. ] [0.8 0. ] [1. 0. ]] res[4]= [0.8 0. ] res[5]= 1.0 ÂüÁ¶) 1Â÷¿øÀ¸·Î MinMaxScaler »ç¿ëÇϱâ numpy reshape¸¦ ÀÌ¿ëÇØ Â÷¿øÀ» ¹Ù²Ù¸é °¡´ÉÇÏ´Ù. reshape¿¡ -1À» »ç¿ëÇϸé numpy¿¡¼ ¾Ë¾Æ¼ º¯È¯ÇÑ´Ù. from sklearn.preprocessing import MinMaxScaler
import numpy as np data = np.array([0, 10, 20, 30, 40, 50]) data = data.reshape(-1, 1) print(data) scaler = MinMaxScaler() res = scaler.fit(data) print('----------------------') print('min=', scaler.data_min_) print('max=', scaler.data_max_) print('------ transform --------') res = scaler.transform(data) print(res) print('res[4]= ', res[4]) print('res[5]= ', res[5][0]) °á°ú) [[ 0] [10] [20] [30] [40] [50]] ---------------------- min= [0.] max= [50.] ------ transform -------- [[0. ] [0.2] [0.4] [0.6] [0.8] [1. ]] res[4]= [0.8] res[5]= 1.0 ---------------------------------------------------------------------------------------------------- reshape(-1, 1)À» »ç¿ëÇÒ¶§ data´Â ´ÙÀ½°ú °°ÀÌ ¹Ù²ï´Ù. [[0. ] [0.2] [0.4] [0.6] [0.8] [1. ]] reshape(1, -1)À» »ç¿ëÇÒ¶§ data´Â ´ÙÀ½°ú °°ÀÌ ¹Ù²ï´Ù. [[ 0 10 20 30 40 50]] °á°úµµ ÀÌ»óÇÏ°Ô ³ª¿Â´Ù. data = data.reshape(1, -1) [[ 0 10 20 30 40 50]] ---------------------- min= [ 0. 10. 20. 30. 40. 50.] max= [ 0. 10. 20. 30. 40. 50.] ------ transform -------- [[0. 0. 0. 0. 0. 0.]] |