-
Notifications
You must be signed in to change notification settings - Fork 71
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
run the code on coco, but can not get the same results shown in the paper #8
Comments
May I ask you for the detail config of training coco (number of gpus and etc)? |
GPU using 8 cards of V100 , and the commands are the same as your provided in the github. |
I check it again. There is a mistake of my script. I am so sorry. The |
Hi @rock4you , may I ask for some new progress? |
The AP of coco val2017 with 300 epochs in Table 2 of the paper is 42.8, |
As coco dataset is large, the result is reported at the last training epoch without serveral times (I guess the result variance is small on coco). BTW, may I ask for your result? |
Still running, the AP around epoch 240 is 0.430. |
Glad to hear the result. As far as I observe, the open-source pre-trained model is a little better than paper report. |
👍🏻 👍🏻 |
Nice to hear the result. Could you attach more detailed COCO style evaluation result (such like https://gist.github.com/dddzg/cd0957c5643f5656f6cdc979da4d6db1)? |
The last epoch: IoU metric: bbox The highest at epoch 288: IoU metric: bbox |
Hi you mean in the finetune stage or pretrain stage ? Why in the pretrain stage the backbone should freeze ? |
The text was updated successfully, but these errors were encountered: