Input
The input table is iris_test (see SVMDense Examples Input).
SQL Call
SELECT * FROM SVMDensePredict ( ON svm_iris_test AS "input" PARTITION BY ANY ON densesvm_iris_linear_model AS model DIMENSION USING IDColumn ('id') InputColumns ('[1:4]') Accumulate ('id', 'species') ) AS dt ORDER BY id;
Output
id | predict_value | predict_confidence | species |
---|---|---|---|
5 | setosa | 0.834460442660295 | setosa |
10 | setosa | 0.77698179935506 | setosa |
15 | setosa | 0.904620370750784 | setosa |
20 | setosa | 0.831036067464484 | setosa |
25 | setosa | 0.713640318278595 | setosa |
30 | setosa | 0.749760136354252 | setosa |
35 | setosa | 0.766957250093069 | setosa |
40 | setosa | 0.804297108005061 | setosa |
45 | setosa | 0.75080175280448 | setosa |
50 | setosa | 0.808477367597895 | setosa |
55 | versicolor | 0.509380447983795 | versicolor |
60 | virginica | 0.437785512959727 | versicolor |
65 | versicolor | 0.254388995091932 | versicolor |
70 | versicolor | 0.572603245374737 | versicolor |
75 | versicolor | 0.508482636653637 | versicolor |
80 | versicolor | 0.525738899884489 | versicolor |
85 | virginica | 0.643995565386973 | versicolor |
90 | versicolor | 0.459132712526973 | versicolor |
95 | versicolor | 0.436287581706344 | versicolor |
100 | versicolor | 0.39310360317574 | versicolor |
105 | virginica | 0.969991307456827 | virginica |
110 | virginica | 0.949973903066985 | virginica |
115 | virginica | 0.981893518564686 | virginica |
120 | virginica | 0.847012885109085 | virginica |
125 | virginica | 0.902480100166191 | virginica |
130 | versicolor | 0.717661810166786 | virginica |
135 | virginica | 0.884508297150359 | virginica |
140 | virginica | 0.815440611025279 | virginica |
145 | virginica | 0.965899115798588 | virginica |
150 | virginica | 0.866724666351753 | virginica |
The prediction accuracy with the linear model is 90%.