which model you choose to finetune?
RokiaAbdeen opened this issue · comments
you saved two model the last and best modet, which one you used for finetuning? for testing?
The official PyTorch implementation of the paper "Learning by Analogy: Reliable Supervision from Transformations for Unsupervised Optical Flow Estimation".
RokiaAbdeen opened this issue · comments
you saved two model the last and best modet, which one you used for finetuning? for testing?