h-bo
h-bo
> visualize both prediction and ground truth on one image, as: true postive (tp), false postive (fp) and false negtive (fn=miss) ~ for example, use different color to draw boxes...
@bhadresh74 I met this issue when using tsn-caffe. In fact, I found the reason was that the caffe version tsn used was different from the latest, and was modified by...
@pedro-abreu I upload my code in my github. But its result is worse than expected. The author of tsn recommend me to use their pyTorch version. Hope to help you.
> 由于qwen-vl默认处理的image token长度为256，如果链接过长会出现这种错误，可以转成短链接 This is inconvient, can it be fixed?