The training list and the values of some parameters

RL4M / PCRLv2

An official implementation of PCRLv2 (pre-training and fine-tuning code are included).

MIT License

115 stars 8 forks source link

The training list and the values of some parameters #4

Open zheangh opened 1 year ago

zheangh commented 1 year ago

Thank you for the excellent work and for releasing it. I wonder how the train txt is generated for LUNA, because in the code you already divide the train/val set by different "subset". In addition, I notice some hyperparameters are different between the paper and the code (e.g., batch size for 2D, weight decay), and the "patience" parameter in finetuning code is missing. Could you confirm their values? Thanks!

liaochuanlin commented 1 year ago

@zheangh 我也发现了这个问题请问你解决了吗？

Fym68 commented 1 month ago

How did you solve the problem with the patience parameter? I also have this problem.