wsy-yjys
wsy-yjys
When I test the fps of edgenext_x_small and edgenext_small with RTX 2060（notebook）, find edgenext_small is faster than edgenext_x_small ？ the following is the result . Each row is the average...
``` xmin = int(get_and_check(bndbox, "xmin", 1).text) - 1 ymin = int(get_and_check(bndbox, "ymin", 1).text) - 1 ``` why '-1' in state which convert xmin and ymin
### Prerequisite - [X] I have searched [the existing and past issues](https://github.com/open-mmlab/mmyolo/issues) but cannot get the expected help. - [X] I have read the [FAQ documentation](https://mmyolo.readthedocs.io/en/latest/faq.html) but cannot get the...
### Prerequisite - [X] I have searched [the existing and past issues](https://github.com/open-mmlab/mmyolo/issues) but cannot get the expected help. - [X] I have read the [FAQ documentation](https://mmyolo.readthedocs.io/en/latest/faq.html) but cannot get the...
```python class OREPA_LargeConvBase(nn.Module): def __init__(self, in_channels, out_channels, kernel_size, stride, padding, dilation=1, groups=1, deploy=False, nonlinear=None): …… internal_channels = out_channels ``` 代码中为什么将3x3conv之间channel设置成与输出一致，而不是输出的倍数，进一步增加参数量，提升模型性能？
1. 下图在验证重参数化前后输出差异的时候，为什么需要先将BN权重初始化？  2. 如果dbb_1x1_kxk中将1x1个数增加（即纵向堆叠很多个1x1，最后一个是3x3），重参数化后误差是否会加大? 期待您的回复~
 Is there a mistake here? The figure is 7x7, not 13x13？
I already use pytorch1.8.0, but still encounter the bug during training, could you give me some help? Thank you ```bash /pytorch/aten/src/ATen/native/cuda/Loss.cu:102: operator(): block: [301,0,0], thread: [60,0,0] Assertion `input_val >= zero...
Why not use cocoapi for evaluation, as far as I know there are differences between cocoapi evaluation results and ultralytics evaluation results
Question 1: Will the batch size during training affect the final performance of the model? Have you tried small batch size, such as 16 and 32? Question 2: Why do...