A Fast Logdet Divergence Based Metric Learning Algorithm for Large Data Sets Classification

Table 1

Data sets used in the experiments.

Name	Number of attributes	Number of classes	Number of instances
Iris	4	3	150
Seeds	7	3	210
Wine	13	3	178
WDBC	30	3	569
Sonar	60	2	208
Ionosphere	3	2	351
Letter-Recognition	16	26	20000
Spambase	57	2	4601
Isolet	617	26	7797
Semeion	256	10	1593
Mfeat	649	10	2600