In pascal_voc.py `class pascal_voc(imdb): def init(self, image_set, year, devkitpath=None): imdb.init(self, 'voc' + year + '_' + image_set) self._year = year self._image_set = image_set self._devkit_path = self._get_default_path() if devkit_path is None \ else devkit_path self._data_path = os.path.join(self._devkit_path, 'VOC' + self._year)

first split

    if cfg.TRAIN.META_TYPE == 1:
        self._classes = ['__background__'] + cfg.TRAIN.ALLCLASSES_FIRST
    #second split
    if cfg.TRAIN.META_TYPE == 2:
        self._classes = ['__background__'] + cfg.TRAIN.ALLCLASSES_SECOND
    #third split
    if cfg.TRAIN.META_TYPE == 3:
        self._classes = ['__background__'] + cfg.TRAIN.ALLCLASSES_THIRD`

you always use ALLCLASSES, and dosenot filter class in phase 1, so you use base+novel class for training RPN of FasterRCNN?

yanxp / MetaR-CNN

Data use error for RPN of FasterRCNN in phase 1 #34

first split