Xiangchao Yan
Hello, when I run DIP to inpainting or sr, the same training parameters often get different PSNR values, and they may differ by more than 2 points. In addition, during...
预测得分为负数
作者您好，我在使用您的代码训练自己的数据时，数据一共有6类，预测结果没有问题，准确率很高，但是其他的类得分为什么会出现负数呢？ 并且不同的图像得分的大小程度也差很多，请问是什么原因呢
Hello, I want to know that styleflow can artificially generate faces with various poses and expressions. How do I use code to generate a series of continuous poses? Which parameters...
Two models with different structures get different confidence**for the same dataset**. For example, most of the confidence obtained by the first model are more than **0.9**, while most of the...
**When the target frame is narrow and long, the regression loss is 0**！ `D:\YXC\Study\Anaconda\envs\pt-gpu\python.exe C:/Users/59593/Desktop/pytorch-retinanet-master/train.py CUDA available: True Num training images: 2 CUDA available: True CUDA available: True CUDA available:...
Hello, why set the pre-trained model to 20 and 10 for waymo and nuscenes, is there any basis for this? And what is the pretrained epoch for KITTI?
Hello, it seems ok when I pretrained the model on KITTI, but the cost is too huge when I pretrain on Waymo, the speed is too slow.
Hello, is your total batch is 4*4 on 4 V100 GPUs? Have you tried bigger batch size? I have tried bigger batch size but the results are not good.
 设定的时间好像没有触发，action里也没有任务
Hi, I have read your paper carefully and the target objects are mostly structures like cells and chips. I have currently photographed a transparent lens cell using traditional LFM and...