Open wangxihao opened 1 year ago
Hi @wangxihao , Thanks for your interest in our work. There is no training step in this method, our optimization takes as input predictions from off-the-shelf models pre-trained on other tasks. The full code is already uploaded.
Thank you for kind explanation.
Hi, it is great work and I just want reproduce the training , wil you plan to release train code, Thank you?