-
I have measured detection speed on Jetson Xaiver NX.
The results is below (using fp16)
nano : 0.0721[s]
tiny : 0.0662[s]
Would you tell me the cause of the result?
Thank you in advance.
-
-
I followed the steps to convert my PyTorch-trained model to onnx-simplifier, then converted that model to ncnn and modified the focus layer. However, in the output section at the bottom, there is no r…
-
我使用0.1.0版本yolox_nano训练自己的数据集，用官方的tensorrt C++部署时推理结果不对，这是什么原因导致![Uploading tensorrt.jp…
zzzfo updated
2 years ago
-
First， Thank you for your team's research and contribution！
Have you tested the results of the free anchor like YOLOX, thx!
-
作者你好！我用你们的yolox训练自己的数据集时发现total_loss一直震荡下不去，我使用的数据集一共有3000张图片，训练集验证集的划分比例为8：2。请问需要进行哪些改动才可以让total_loss降下去呢？谢谢！
![1](https://user-images.githubusercontent.com/53384442/127726135-9482046d-ee64-4750-beb…
-
As mentioned in title, have you used giou/diou/ciou to train yolox, and who is better?
-
li@li-desktop:~/Desktop/YOLOX-main$ python3 tools/demo.py image -f exps/default/nano.py -c yolox_nano.pth --path assets/dog.jpg --conf 0.25 --nms 0.45 --tsize 640 --save_result --device gpu
T…
-
Hi,
I have 2 bugs related to slicing.
1. In the `forward` method of a torch module, that uses rank 4 tensors, attempting to slice and assign to a number.
2. In the `forward` method of a torch m…
-
您好，我做了以下改变，将主干网络换为mobilenetv3，加载预训练模型，epoch为100，其他参数没变，在云端GPU训练，不管是冻结训练还是解冻训练，测试结果准确率召回率均为0，map值极低，百分之个位数。检查了一下，model_path，classes_path，classes.txt均改为自己的数据，图片路径和标签也没错，损失也降到了零点几左右，但是检测结果感觉跟没训练一样，不知道是哪方…