ijkguo / mx-rcnn

Parallel Faster R-CNN implementation with MXNet.
Other
669 stars 292 forks source link

Difference between CXX_PROPOSAL and custom proposal? #61

Closed xinghedyc closed 7 years ago

xinghedyc commented 7 years ago

Hi, thanks for your great work, it really helps me a lot, However I don't quite know about the difference between CXX_PROPOSAL and custom proposal. as in symbol file:

if config.TRAIN.CXX_PROPOSAL:
            rois = mx.contrib.symbol.Proposal(
            cls_prob=rpn_cls_act_reshape, bbox_pred=rpn_bbox_pred, im_info=im_info, name='rois',
            feature_stride=config.RPN_FEAT_STRIDE, scales=tuple(config.ANCHOR_SCALES), ratios=tuple(config.ANCHOR_RATIOS),
            rpn_pre_nms_top_n=config.TRAIN.RPN_PRE_NMS_TOP_N, rpn_post_nms_top_n=config.TRAIN.RPN_POST_NMS_TOP_N,
            threshold=config.TRAIN.RPN_NMS_THRESH, rpn_min_size=config.TRAIN.RPN_MIN_SIZE)
else:
            rois = mx.symbol.Custom(
            cls_prob=rpn_cls_act_reshape, bbox_pred=rpn_bbox_pred, im_info=im_info, name='rois',
            op_type='proposal', feat_stride=config.RPN_FEAT_STRIDE,
            scales=tuple(config.ANCHOR_SCALES), ratios=tuple(config.ANCHOR_RATIOS),
            rpn_pre_nms_top_n=config.TRAIN.RPN_PRE_NMS_TOP_N, rpn_post_nms_top_n=config.TRAIN.RPN_POST_NMS_TOP_N,
            threshold=config.TRAIN.RPN_NMS_THRESH, rpn_min_size=config.TRAIN.RPN_MIN_SIZE)
ijkguo commented 7 years ago

None except C++ or Python.

xinghedyc commented 7 years ago

Ok, Thank you.