ailias/Focal-Loss-implement-on-Tensorflow
The implementation of focal loss proposed on "Focal Loss for Dense Object Detection" by KM He and support for multi-label dataset.
PythonApache-2.0
Issues
- 1
Focal loss with MobileNet-SSD FPNLite.
#16 opened by Annieliaquat - 2
∵0**0=1 ∴ gamma != 0
#15 opened by zhangyuqi-1 - 0
因为0**0=1,所以需要增加γ=0时的特殊情况。否则会出现问题
#14 opened by zhangyuqi-1 - 0
多分类时参数α的意义是什么?
#13 opened by ChuanTianML - 6
Focal loss value is very less
#2 opened by prakashjayy - 2
The Code can be try in multi label ?
#8 opened by hecongqing - 0
train my data
#12 opened by henbucuoshanghai - 0
tarin my data???????how???????
#11 opened by henbucuoshanghai - 4
- 1
Why add "neg_p_sub" into "per_entry_cross_ent"
#10 opened by CoderChang - 7
- 4
How to implemented it in faster-rcnn?
#6 opened by zqdeepbluesky - 1
what does array_ops.where mean?
#7 opened by DRosemei - 1
tf.nn.sigmoid and tf.nn.softmax diff?
#4 opened by ronyuzhang - 1
The issue about @slim.add_arg_scope
#3 opened by zzzzzz0407 - 1