Loneseven issues

Repositories
Issues
Comments

Results 1 issues of


                                            Loneseven

About the pre-train strategy in the code

At the moment the internvl_chat/internvl/train/internvl_chat_finetune.py and internvl_chat/internvl/train/internvl_chat_pretrain.py are almost the same. I am wondering if the ITM loss and Contrastive loss would be used. Am I miss something? Is generative...