icyzhang

hi~ in your pre-trained models, the name of "aj_rgb_charades.pth", it means "aj_rgb_kinetics.pth" or a finetuning I3D model on Charades dataset

I have another problem, when I test the aj_rgb_charades.pth model, the mAP is 24.8%. If I train just one epoch(lr 0.1), the mAP is 32.9%. And I notice that the...

Thank you for your reply!

As you describe in another issue The learning rate was set to 0.1, then dropped to 0.01 after 45k steps. The batch size was set to 6 videos per-GPU, an...

> This problem is a multi-label classification problem and does require BCEloss. But I don't know why the training convergence is very slow. I use 4*1080Ti at a time, 64...

I think it's may be caused by the hyper-parameters settings and the numbers of gpus

you need to compile the matconvnet again under the condition more than one gpu~try it~

I have another problem about the two-stream CNN,when I compiled MexConv3D, there is an error'CL.EXE not found in PATH. Trying to guess out of mex setup. 'I don't know why

I have the same question,have you solved it?