Closed Kyfafyd closed 2 years ago
If you use a batch size of 16, but let the gradients accumulate for 2 batches (grad_ac_step = 2), then it uses an effective batch size of 32. It helps when you can't store all 32 images in GPU memory
Dear author,
Thanks very much for you response!! I am clear now. I have another question: which type and how many gpus are you used for training on COCO and VOC?
We used four NVIDIA Titan X GPUs
Dear author,
Thanks very much for you patience! How long it will take for COCO and VOC, respectively? It seems it is very slow for COCO...
VOC: ~5 minutes per epoch COCO: ~1 hour per epoch
But it seems the default training config is 100epochs, so when the model will be converged, especially for COCO? Until 100 epochs?
It's possible it could take that long. you will have to run it for at least a few days.
On Mon, Nov 8, 2021 at 10:38 AM Zhao WANG @.***> wrote:
But it seems the default training config is 100epochs, so when the model will be converged, especially for COCO? Until 100 epochs?
— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/QData/C-Tran/issues/11#issuecomment-963281291, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJFX7C2ORI5Z6FQ3LTXG43UK7VIJANCNFSM5HSWA4XA .
So, have you used 4 GPUs with a batch size 16? It seems 4 GPUs is capable for a batch size 32.
Yes, you could do that
On Mon, Nov 8, 2021 at 10:44 AM Zhao WANG @.***> wrote:
So, have you used 4 GPUs with a batch size 16? It seems 4 GPUs is capable for a batch size 32.
— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/QData/C-Tran/issues/11#issuecomment-963286731, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJFX7CSIIEBX3L6EUAYL43UK7V6RANCNFSM5HSWA4XA .
Thanks very much for your response!
Excuse me, did you get the mAP of voc 2007? It was not mentioned in the paper.
We did not run experiments on voc 2007 as it is a small dataset with very few labels.
Dear author,
Thanks very much for your interesting work! I wonder why the args "--grad_ac_step" is set as 2 for VOC2007?
From your code, I see that if you set "--grad_ac_step" as 2, a minibatch images (16 images) will not be used for optimization. Can you help me solve this issue?
Thanks very much!