VideoTextSCM

VideoTextSCM copied to clipboard

totaltext_ Resnet50 Baidu Netdisk

2

totaltext_ Resnet50 Baidu Netdisk cannot be downloaded. Can you share it again？Thanks!

yywwwwww

https://github.com/lsabrinax/VideoTextSCM/blob/d87ad1bbb6ada7573a02a82045ee1b9ead5861ad/train_embedding.py#L47 Thankyou for share the code! But what's the mean of 'W_scale' in spatical_triplet_loss, and loss_pos = pos_dist * (W_scale + W_pos_dis) - alpha1, why we need to add 'w_scale'?

sheirving

Hi, thank you for your hard work with the model. Will you guys release the pretrained model? Thanks

joeyism

请问在测试的时候有遇到Attempted to set the storage of a tensor on device “cuda:0“ to a storage on different device “cpu“.这个问题吗？这是由于什么引起呢，是我的torch1.13版本不匹配吗

MxxM-max