Skip to main content
Fig. 13 | Journal of NeuroEngineering and Rehabilitation

Fig. 13

From: Cybathlon experiences of the Graz BCI racing team Mirage91 in the brain-computer interface discipline

Fig. 13

2D representation of the 48-dimensional feature space computed with t-SNE: Each point corresponds to an observation. The smaller the distance between the 2D points, the closer they are in the 48D feature space. (left) Data of the six calibration runs (crosses) on which the sLDA classifier was trained on. The training labels were used to color-code the 2D points. The other plots additionally summarize observations during games in the container (middle, circles) and the arena (right, squares). We used the sLDA classifier output to color-code the points for this data

Back to article page