question about masking

[arslan-rashidov]

Hey there! I want to pretrain the model on my custom dataset from scratch. But my model wasn't learning until I have made dynamic masking of series. My question is - should I dynamically mask only train set(mask series in getitem)? And valid set should be static for evaluation of training process, right? Correct me please, if I've mistaken

[gzerveas]

Hi, can you give a few details about what you are trying to achieve? Are you interested in imputation? If so, maybe this post will help you: #41 (comment)