Open Oufattole opened 1 month ago
The token loss will be the same as for forecasting: https://github.com/Oufattole/meds-torch/blob/52ed2fbac72fb72a6f9491339cded5ba10354a2f/src/meds_torch/models/token_forecasting.py#L138
Lemme know what you think @teyaberg
Implementation of a Masking Stage with Random Masking Options
Problem
The absence of a dedicated masking stage in our pipeline limits our ability to handle incomplete or noisy data effectively during model training.
Proposed Solution
Introduce a masking stage designed to randomly mask a specified percentage of the data or subsequences within the data: