Closed johntiger1 closed 3 years ago
Hi, thank you for your interest! I installed new virtual environment for detection and the command to extract features is available at https://github.com/uclanlp/visualbert#extract-image-features-on-your-own.
Hi @liunian-harold-li, @johntiger1, how is the speed of extracting image features?
For example, by comparison with one GPU (cpu-only is presumably not tolerable), for NLVR2 107,292 images, lxmert takes 5-6 hours to extract faster-rcnn features by this caffe.
https://github.com/uclanlp/visualbert#extracting-image-features
Could you go into more detail? Should we install the custom pytorch into a new virtual environment, so it doesn't break the pytorch used in training the model? What command do we run with detectron to extract features?
Hi @johntiger1, if you have not finished this, i am interested in figuring this out as well recently.
Hi @liunian-harold-li, @johntiger1, how is the speed of extracting image features?
For example, by comparison with one GPU (cpu-only is presumably not tolerable), for NLVR2 107,292 images, lxmert takes 5-6 hours to extract faster-rcnn features by this caffe.
Hi @liunian-harold-li, @johntiger1, I get my solution for this question of time estimate and summarize it here. Thanks anyway.
Thank you for testing the time estimate!
I find you are all talking about this url: https://github.com/uclanlp/visualbert#extracting-image-features while it is a same web of https://github.com/uclanlp/visualbert when I open it, what happened and could anyone tell me something about https://github.com/uclanlp/visualbert#extracting-image-features?
https://github.com/uclanlp/visualbert#extracting-image-features
Could you go into more detail? Should we install the custom pytorch into a new virtual environment, so it doesn't break the pytorch used in training the model? What command do we run with detectron to extract features?