yolov2转caffe模型中出现了检测错误错误

Question

yolov2转caffe模型中出现了Check failed: top_shape[j] == bottom[i]->shape(j) (26 vs. 13) All inputs must have the same shape, except at concat_axis.错误

Answer 1

28层输入维度不匹配，你的25层是[1,1024,13,13],26层是[1,512,26,26]
具体的修改方案你可以看看下面这个能不能用，不行的话就得自己对25层进行一次reshape或者卷积之后和26层一样才可以进入下面一层（建议是reshape，因为再次卷积可能会导致数据变换）

Convert darknet yolov2 model to caffe · Issue #24 · ysh329/deep-learning-model-convertor · GitHub I tried pytorch-caffe-darknet-convert but failed due to lack of reorg layer and issue of concate layer support (of course, I created an issue to ask but no one replied me). Next I think three available solutions below: try to write code ... https://github.com/ysh329/deep-learning-model-convertor/issues/24

如果上面的方法还是不行的话，你需要考虑下转成其他模型，类似ONNX看看？