You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello,
I am really interested in trying out your framework. Could you elaborate a little bit more on the installation process please?
I did install Detectron2 and Fast-Transformer, but what should I do with DETR and Deformable-DETR, as you can't really install those as packages to your environment? Are these supposed to be added as submodules?
Thank you for your time.
The text was updated successfully, but these errors were encountered:
you can obtain similar behavior by multiplying the weights of the self attention module by a binary mask where 1 correspond to the 100 nearest keys for each query. To train, don't forget to first convert you json using our provided code, then you can simply run train_net and specify the number of gpu and the path for you config for instance python $ROOT/train_net.py
--num-gpus $NUM_GPUS
--config-file $ROOT/configs/file.yaml
Hello,
I am really interested in trying out your framework. Could you elaborate a little bit more on the installation process please?
I did install Detectron2 and Fast-Transformer, but what should I do with DETR and Deformable-DETR, as you can't really install those as packages to your environment? Are these supposed to be added as submodules?
Thank you for your time.
The text was updated successfully, but these errors were encountered: