The idealization i.e. removing of outliers
Method is taken in Method.txt
Data from #TrainingSet.txt
Molecular representation is  SMILES
Correlation coefficient (test set) with all outliers    0.7885
Standard Error of Estimation (Test set)    0.5894

Status of test set after the removing of "worst" outliers:

 Number of  ID of          R2      dR2       SEE      Number of
 outliers   outlier                                   compounds
    0  :               : 0.7885:   0.0000:   0.5894:      101
    1  :            701: 0.8005:   0.0120:   0.5679:      100
    2  :            385: 0.8106:   0.0101:   0.5538:       99
    3  :            118: 0.8189:   0.0083:   0.5456:       98
    4  :            405: 0.8249:   0.0060:   0.5349:       97
    5  :            355: 0.8311:   0.0061:   0.5236:       96
