This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
I hope the andonios figured it out (after inserting the missing space before "-h"). I had the same issue, and my problem was with backup files of opencv_traincascade and with varying sample size.
opencv_traincascade saves additional .xml files "for the case of interrupted training". First, I created a .vec file using one pair of values for sample size and trained the model. Then I decided to use a different sample size. If I didn't remove the opencv .xml backup files, opencv_traincascade would throw the error described in the question. I guess, because it decides to load those files if they exist.
If this your case, simply deleting all opencv .xml files before another training should help.
| | 2 | No.2 Revision |
I hope the andonios figured it out (after inserting the missing space before "-h").
I had the same issue, and my problem was with backup files of opencv_traincascade and with varying sample size.
opencv_traincascade saves additional .xml files "for the case of interrupted training". First, I created a .vec file using one pair of values for sample size and trained the model. Then I decided to use a different sample size. If I didn't remove the opencv .xml backup files, opencv_traincascade would throw the error described in the question. I guess, because it decides to load those files if they exist.
If this your case, simply deleting all opencv .xml files before another training should help.
