SDDRAWROC Draw interactive ROC plot FH=SDDRAWROC(R,COLSTR) INPUT R SDROC object COLSTR cell array with color, line and marker strings (see PLOT) OUTPUT FH figure handle FIGURE INTERACTION Cursor keys Change performance measures w show/hide operating point data (weights/threshold) s save current op.point to a pipeline or algorithm f fold info: Toggle between error bars and per-fold realizations (only for multi-fold ROC, see SDCROSSVAL) DESCRIPTION SDDRAWROC plots an interactive 2D ROC plot. Moving the mouse, different operating points may be investiagted. Pressing the 'w' key toggles display of weights/threshold in the figure title. Different measure combination may be selected using the cursor keys. Clicking on any operatin point makes it current (black). Using 's' key, operating point may be saved back in Matlab workspace into existing ROC object, pipeline or SDALG algorithm. EXAMPLES Perform ROC analysis for sdlinear classifier: >> [tr,ts]=randsubset(a,0.5); >> p=sdlinear(a) >> out=ts*p >> r=sdroc(out) >> sddrawroc(r) READ MORE http://perclass.com/doc/guide/ROC_analysis.html#intro SDROC
sddrawroc is referenced in examples:
- kb24: Example on building an image detector
- kb19: PRTools compatibility
- kb17: How to optimize three-class classifier in imbalanced problems
- kb10: A step by step construction of a detector
- kb9: How to build a detector from a custom region in an image?
- kb8: How to build a detector in a single line of code?
- kb6: Can ROC analysis be performed for multi-class problems?
- kb5: ROC analysis on two-class problems: choosing an operating point