Alex Hu

hello haomood, i still face the same problem as yours, i use video from HMDB-51 dataset (originally .avi, i use ffmpeg filename.avi filename.mp4 to convert to mp4 video), but i...

Can you explain the specific settings during your test?

Thanks for your suggestion. They should be both in [-1,1] range.

_CLIP_SIZE means in the #num of frames you select for each video during training; _EACH_VIDEO_TEST_SIZE means in the #num of frames you select for each video during testing; _FRAME_SIZE means...

clip size is not the same as the number of frames put in the txt document, it is the # sampling frames, for details you should refer to the paper...

can you inform me of your learning rate? and at first, the accuracy may be not stable, but when the training continues, it may relieve.

It seems you face out of memory of GPU. We run our code on GTX-1080Ti (12GB).

hi panna, you can reduce the batchsize(BATCH_SIZE) or the frame rate(_CLIP_SIZE ) in finetune.py.

Thanks for your reply! I have solved the environmental work. It is quite tricky and needs to be careful at every step~ If having time, I may write a blog...

Hi，ARUNIRC， would you release pre-trained model now ? Thanks in advance !