The analysis of an opportunity of classification of primary nuclei of cosmic rays using space-temporary distributions of Cherenkov light air shower is carried out. For this aim the full information of space-temporary distributions was used. The method of Support Vector Machine was used for classification. This technique allows realizing good classification for three-component model of cosmic rays. The probability of true classification gets ∼60÷80% for O and H nuclei and ∼80÷100% for Fe nuclei.