chuanqi305

@zhanglonghao1992 Focal loss is not necessary in two-stage detection frameworks. The class imbalance problem only exists in one-stage detection networks such as YOLO and SSD.

@mychina75 I found the error and corrected it, for verification I checked the gradient by check_focal_diff.py. Sorry for my fault, please check out the new code and test. @bailvwangzi I...

@mychina75 What's the final loss value? In my test the evaluation is OK.

@bailvwangzi No, I did not get a higher mAP, too. Just the same as OHEM.

@mychina75 Too few iterations, you should evaluate after iteration 30000~50000.

@XiongweiWu Can you talk about some details? In my test, the performance has not been improved, focal loss is not better than OHEM.

No, the loss should be < 10 after 10 iterations. Maybe there is a bug in your network structure?

@pbdahzou Maybe the Focal Loss is similar to OHEM in the training effect. The retinaNet use FPN framework, maybe the key factor is 'Deconvolution'.

@xiaorannuo No third-party framework used, just implemented with RenderScript and java.

@contactmat85 You should merge the batch norm layers into conv layers， use my merge_bn.py in mobilenet-ssd project to do that.