sound

You can add it in config file (configs/fcos/fcos_R_50_FPN_1x_center_giou.yaml) @18804601171

In my experiment, 1.5 is the best. @18804601171

Gt will be implicitly expanded in this line, I just expanded it in advance. @18804601171 https://github.com/yqyao/FCOS_PLUS/blob/0d20ba34ccc316650d8c30febb2eb40cb6eaae37/maskrcnn_benchmark/modeling/rpn/fcos/loss.py#L66

I don't understand your question very well, I think all gt will be passed into function to get sample region. Maybe you need to read our recent paper https://arxiv.org/abs/1912.02424. @18804601171

At the begin of training, it's ok. After several epochs, the loss will decrease. @lingtengqiu

you need change NUM_CLASSES in configuration.yaml @haochange

readme 已经说了，就是把原始的VOC2007 VOC2012 ImageSet/Main/train.txt val.txt 等等 直接cat 合成一个大的 @ @yongqiang1210

你在代码对应位置打印一下信息，看看是不是你的list写错了 @yongqiang1210

How about your training loss ? @pengjizhi I think the value of your training loss may be very large.

What dataset you use for training ? @pengjizhi