Open Oufattole opened 2 months ago
follow the hydra target schema: https://github.com/Oufattole/meds-torch/blob/3533dfadaf472b4cd2a01bb5c25a723f6d92be22/configs/model/ocp.yaml#L7
priority tag specifically commenting on comparing ways to turn a coninuous value (code-specific or code-independent) into a vector embedding (e.g., scaling a code-independent embedding vector, scaling a code-specific embedding vector, using an FFN, etc).
We currently only support Continuous value embeddings (a one to many FFN). We should try other things, like supporting quantizing.