[Deep learning] [Original] Using darknet yolov3 to train yolov3-spp pre-training model acquisition method

The training process of yolov3-spp is exactly the same as that of yolov3, but where is the pre-training model? I can find it on the Internet. You can use the following method to generate

./darknet partial cfg/yolov3-spp.cfg Yolov3-spp.weight yolov3-spp.conv.15 15

Update note: The pre-training model generated by the above is only 3.6M, which should not work. Later, the article asked, just use the yolov3 pre-training model. Then I tested the yolov3 pre-training model and found that there is no problem. Use yolov3 to pre-train the model.