Check failed: top_shape[j] == bottom[i]->shape(j) (24 vs. 26) All inputs must have the same shape, except at concat_axis

ChenYingpeng / darknet2caffe

Convert darknet weights to caffemodel

183 stars 88 forks source link

Open l1uw3n opened 4 years ago

l1uw3n commented 4 years ago

joel5638 commented 4 years ago

@l1uw3n any luck resolving the issue? I have the same issue. Can you please guide

silverlining21 commented 4 years ago

pooling layer parameter mismatch

Wangxiangyao-717 commented 3 years ago

I meet the similar issue when I use it to change yolov4. At last, I find my yolov4.cfg lose 'pad' in 'maxpool'. You can try to check this. Good luck.