Edge Impulse Experts / Glass break detection with Rubik PI 3 Public
Set the 'expected outcome' for each sample to the desired outcome to automatically score the impulse.

Apply filters

Sample name Expected outcome Length Accuracy Result
204777__ngruber__breaking-glass.s4 glass 3s 80% 4 glass, 1 uncertain
387349__cosmicembers__powerful-crush glass 3s 20% 4 uncertain, 1 glass
396289__felix-blume__glass-hit-impact.s8 glass 3s 60% 3 glass, 2 uncertain
396289__felix-blume__glass-hit-impact.s16 glass 3s 40% 3 uncertain, 2 glass
204777__ngruber__breaking-glass.s1 glass 3s 80% 4 glass, 1 uncertain
619039__mrrap4food__breaking-glass-1.s6 glass 3s 100% 5 glass
607910__ienba__glass-bottle-breaking.s3 glass 3s 100% 5 glass
591069__the-epic1357__breaking-glass.s1 glass 3s 100% 5 glass
street3.s1 street 3s 0% 5 uncertain
street3.s8 street 3s 0% 5 uncertain
Filter yielded no results, click to adjust the filter

Model testing output

Results

Model version:
Area under ROC Curve 0.89
Weighted average Precision 0.94
Weighted average Recall 0.94
Weighted average F1 score 0.94
Loading...
Loading...