Fig. 10: Iris data set.
From: Nearest centroid classification on a trapped ion quantum computer
a Ratio of distance calculated from the quantum computer vs the simulator, and b classification of Iris data. There were 150 data points in total. The dashed line shows the accuracy of classical nearest centroid algorithm. The error bars in a correspond to the standard error in the ratio of distances for each dataset. Iris data classification pictured after using principal component analysis to reduce the dimension from 4 to 2. The color of the boundary of the circles indicates the three human-assigned labels. The color of the interior indicates the class assigned by the quantum computer. c shows the classification before error mitigation using 500 shots, and d shows the classification after.
