Page 135 - Artificial Intellegence_v2.0_Class_12
P. 135

esidual
                           arks   )                                                             rror









                                      1           3                 6                 9
                                                    o. of  ours  tudied   )

             o,      is calculated as
                                                          N
                                                         ∑  (Pr edicted − Actual ) 2
                                                                             i
                                                                     i
                                                RMSE =    i=1
                                                                   N
             he errors are squared before being averaged in     .  his basically means that      gives larger mistakes a higher
             eight.  his suggests that      is far more beneficial  hen substantial errors e ist and have a significant impact on
            the model's performance.  his characteristic is important in many mathematical calculations since it avoids taking the
            absolute value of the error.  he      of a good model should be less than    .  he lo er the      value, the higher
            the model's performance.

                                                                                                Experiential Learning

                        Video Session


                   can the    code or visit the follo ing link to  atch the video   hat is  oot  ean  quare  rror      )
                  https      .youtube.com  atch v   y  qdI as
                  After  atching the video, ans er the follo ing question
                   hat do you mean by










            C alculating  R MS E in Python
            imp or t nu mp y  as np
            y _ p r ed  =  np .ar r ay ( [ 0 .0 0 0 ,   .1 6 6 ,   .3 3 3 ] )     predicted values
                                           0
                                                   0
            y _ tr u e =  np .ar r ay ( [ 0 .0 0 0 ,   .2 5 4 ,   .9 9 8 ] )    actual values
                                           0
                                                   0
            d ef  r mse( p r ed ictions,  tar gets) :
                d if f  =  p r ed ictions -  tar gets
                d if f _ sq  =  d if f  * *  2





                                                                                       C apstone  P roj e ct  133
   130   131   132   133   134   135   136   137   138   139   140