svm_results.txt

Average cross validation score for SVM is 96.49199999999999%.
Number of support vector for validation set 0: 4873
Confusion matrix for validation set 0:
[[969   7  10   1   1   4]
 [  4 975   6  30   1   1]
 [  5   3 940   2  12   2]
 [  5  22   4 957  10   4]
 [  9   0  13  10 990   3]
 [  0   0   0   0   0   0]]
Number of support vector for validation set 1: 4861
Confusion matrix for validation set 1:
[[973  13   6   1   3   2]
 [  8 982   1  23   0   6]
 [ 11   0 978   2  11   2]
 [  4  25   7 937  17   5]
 [  2   0  15  11 951   4]
 [  0   0   0   0   0   0]]
Number of support vector for validation set 2: 4894
Confusion matrix for validation set 2:
[[ 983   11    4    5    3    5]
 [   7  941    3   37    1    2]
 [  12    0  950    0   22    2]
 [   3   29    4  932   11    2]
 [   3    1    9    6 1008    4]
 [   0    0    0    0    0    0]]
Number of support vector for validation set 3: 4910
Confusion matrix for validation set 3:
[[ 943    6    3    1    1    0]
 [  10  932    1   37    1    5]
 [   5    0 1034    4   13    8]
 [   9   21    5  958   15    5]
 [   1    0    6   10  964    2]
 [   0    0    0    0    0    0]]
Number of support vector for validation set 4: 4887
Confusion matrix for validation set 4:
[[1029    6    6    0    2    2]
 [  11  950    2   21    1    1]
 [  10    0  956    2   12    2]
 [   4   30    5  950   14    6]
 [  11    0   12   11  941    3]
 [   0    0    0    0    0    0]]
Over all Confusion matrix for SVM:
[[4897   43   29    8   10   13]
 [  40 4780   13  148    4   15]
 [  43    3 4858   10   70   16]
 [  25  127   25 4734   67   22]
 [  26    1   55   48 4854   16]
 [   0    0    0    0    0    0]]