Skip to main content
. 2022 Dec 2;22(23):9450. doi: 10.3390/s22239450
Algorithm 1: Model Performance Evaluation
Input: Denote cellni as the predicted cell number of the target i in data n.
Output: Denote scorei as model evaluation score with the target i.
i: the target index
c: the cell index
n: the index of data samples
K: the cell index set of actual target presence and neighboring cells
K: the number of neighboring cells (i.e., K=Ki)
  • 1:

     ▶ Initialize:

  • 2:

    scoretruei, scoredeviationi, scorei0

  • 3:

    scoretrue1I           ▷ Maximum target number I.

  • 4:

    scoredeviation12×I      ▷ The evaluation scores.

  • 5:

    n0

  • 6:

     ▶ Start:

  • 7:

     While n < the number of data records for model prediction

  • 8:

     ▶ Initial Sequence (target i in location cellnic)

  • 9:

     for 0<I

  • 10:

      for 0 < K (i.e., the cardinality of the set K)

  • 11:

            if cellni is equal to solutionni

  • 12:

              scoretrueiscoretruei+scoretrue

  • 13:

           else if cellni is equal to othernci

  • 14:

      scoredeviationiscoredeviationi+scoredeviation

  • 15:

      else

  • 16:

      scoredeviationiscoredeviationi

  • 17:

      end if

  • 18:

      end for

  • 19:

      end for

  • 20:

      ▶ Permutation Invariant (target i in location cellnj(c))

  • 21:

      ▷ c K and  K =Kc

  • 22:

      for jK

  • 23:

      if scoretruej and scoredeviationj do not increase

  • 24:

      for 0<K i.e., the cardinality of the set K

  • 25:

      if cellnj is equal to solutionnj

  • 26:

      scoretruejscoretruej+scoretrue

  • 27:

      else if cellnj is equal to otherncj

  • 28:

           scoredeviationjscoredeviationj+scoredeviation

  • 29:

      else

  • 30:

      scoredeviationjscoredeviationj

  • 31:

      end if

  • 32:

      end for

  • 33:

      end if

  • 34:

      end for

  • 35:

      nn+1

  • 36:

      end while

“▶” is the comment of main step and “▷” is the comment of note.