liuxingyu123

The effect after training is not as good as the pre training model you gave

1

my use pytorch 1.7, gpu:1, bachsize:4, epochs:70 picture：[https://s3.bmp.ovh/imgs/2022/05/08/1549ef0c29221c39.png](url)

The reproduced results are relatively poor

6

Hello, the pretrained model I use for vit is vit_ large_ patch16_ 384, but the results of the first stage of training are ODS: 0.74116, OIS: 0.761012, AP: 0.736421. In...

multi-scale test have some question

3

Hello author, do you know why the improvement after multi-scale testing is very small?