-
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…
-
I would like to know the fps in video detection of the YOLOX series, from which results can I check?
Thank you in advance.
-
## 背景
PyTorch 最近在 2.1.0 版本发布了 `torch.export` 功能,提供了以 PyTorch 2.x 的 dynamo 为核心的模型导出方法。其功能上与我们 Paddle 动转静模型导出 `paddle.jit.save` 相类似,因此我们想要建设一个 repo 能够自动对比 `torch.export` 和 `paddle.jit.save` 的导出成功与否,[…
-
输入命令刚开始执行,显卡的内存就被占满了,有时候在第一轮出错,有时候在第三轮,
num_works 设置了4,2,0都是同样的问题,
环境
python 3.8
pytorch1.8.1
cuda10.1
==============================================================================
2021-…
-
Respected sir,
I have purchased this asset but its YOLO sample is very slow on mobile devices. Is there any way to run it on Mobile phone GPU ?
Thank you
-
之前训练yolox_s的时候还好,训练出来的parameter是8.95M,和论文里面的9M正好匹配。可惜觉得速度还是慢了,在cpu下只有7帧的效果,所以想试下最小的nano,可是训练完后发现parameter有2.25M,Gflops2.91。和论文里面的0.91M差距很大,我就把yolox_voc_s里面的width和depth改成了nano的0.33和0.25,然后换了yolox_nano.…
-
训练几个epoch后出现torch.multiprocessing.spawn.ProcessExitedException: process 1 terminated with signal SIGABRT
-
Hi sorry for bothering again!!
I haved tried tiny-yolov4 recently, however, when I convert the weights i got
```bash
TypeError: buffer is too small for requested array
```
The weight and con…
-
请问一下up该怎么批量计算mAP啊,目前用的是固定epoch次数计算mAP感觉随机性比较大。
-
Can you provide the initialization pre training weights used for training YOLOX (YOLOX SwinTiny and YOLOX SwinBase)? I don't have this weight, so I can't get your results by training YOLOX SwinTiny fr…