Nguyễn Duy Cương

VOC 2007 metric?

13

Hi @Cartucho As report here [https://github.com/pjreddie/darknet/issues/956](url) Can we modify our tool to use VOC 2007 metric (Reject difficult objects)?

For more suitable model to deploy in FPGA hardware, i tried to training this model with no bias. But the model converged slowly (mAP~35% after 200k iter with GTX 1080)....

I just run a ssd's model by CHaiDNN and i got this error: > [EO001] Due to mismatch between layer group value and output dimension, Current version support layer group...

Is there anybody successfully build this code in window?

Train network with no bias in convolution layer

10

Hi @zhreshold I already train my mobilenetSSD network in Caffe with no bias. However, the convergence speed of network is too slow (mAP~35% after 3 days) I just try Mxnet...

is there a way to conv from theano's model to caffe?

Hi a Quốc, Tks anh đã public những gì đã đạt được trong repo. Đây là tài liệu rất quý giá cho những người đang thử nhận dạng chữ viết...