search

Chen-Yang-Liu / RSICC

Official LEVIR-CC dataset and Pytorch implementation for Remote Sensing Image Change Captioning With Dual-Branch Transformers: A New Method and a Large Scale Dataset
110 stars 6 forks source link

Could you please share the vocab file to support the pretrained model? #2

Open ShizhenChang opened 1 year ago

ShizhenChang commented 1 year ago

When trying to reproduce the result with the pretrained model, I set the min_word_freq=6 to generate vocab dictionary but obtain a bad performance, so could you share the vocab files that will has consistent results with the pretrained model?

Chen-Yang-Liu commented 1 year ago

You can set the min_word_freq=5 to obtain the all files that can adapt the pre-trained model.