This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
the only classes, that can make use of cv2.ml.STAT_MODEL_UPDATE_MODEL are ANN_MLP and LogisticRegression.
| | 2 | No.2 Revision |
the only classes, that can make use of cv2.ml.STAT_MODEL_UPDATE_MODEL are ANN_MLP and LogisticRegression.
| | 3 | No.3 Revision |
" to get an incremental version of the SVM. " -- sorry, won't work. first thing it does in train() is call clear().
the only classes, that can make use of cv2.ml.STAT_MODEL_UPDATE_MODEL are ANN_MLP and LogisticRegression.LogisticRegression. (also, you would have to make an initial run without that flag)
cv2.ml.STAT_MODEL_RAW_OUTPUT -- that too, goes entirely unused. if you need the distance to the margin, you have to retrieve the support vector(s) using getWeights(), and do your own dot product.
| | 4 | No.4 Revision |
" to get an incremental version of the SVM. " -- sorry, won't work. first thing it does in train() is call clear().
the only classes, that can make use of cv2.ml.STAT_MODEL_UPDATE_MODEL are ANN_MLP and LogisticRegression. (also, you would have to make an initial run without that flag)
cv2.ml.STAT_MODEL_RAW_OUTPUT -- that too, goes entirely unused. if you need the distance to the margin, you have to retrieve the support vector(s) and the shift val using getWeights(), getShift(), and do your own dot product.product, then subtract the shift.
| | 5 | No.5 Revision |
" to get an incremental version of the SVM. " -- sorry, won't work. first thing it does in train() is call clear().
the only classes, that can make use of cv2.ml.STAT_MODEL_UPDATE_MODEL are ANN_MLP and LogisticRegression. (also, you would have to make an initial run without that flag)
cv2.ml.STAT_MODEL_RAW_OUTPUT -- that too, goes entirely unused. if you need the distance to the margin, you have to retrieve the support vector(s) and the shift val using getWeights(), getShift(), and do your own dot product, then subtract add the shift.
