Justin Xu
Justin Xu
Would be incredible if the token count feature was available again (think was implemented in legacy version per #332). Maybe even a total USD cost estimate hidden in the profile...
## Describe the bug The pip version seems to be oudated despite the version number being the same. I checked the code files from the repo vs. the ones installed...
Possible area of enhancement regarding memory usage when a high number of predicates is specified. I remember when I profiled, the memory peaked during the creation of the predicate columns....
Allow derived predicates that are dependent on other derived predicates
Functionality of #42 but usable in aces-cli as opposed to only in Python API
OMOP - [ ] Extraction mode for OMOP CDM - [ ] Demonstration using MEDS OMOP ETL or using mapping from OHDSI ATLAS i2b2 - [ ] Extraction mode for...
- [ ] Add ability to perform case-control matching
Right now we basically only support binary classification. This is ok for now, but we could aim to support more complex tasks: - [ ] Ordinal regression - [ ]...
- [ ] Enable `OR` relationship across branches in the task tree
- [ ] Test LLMs to extract predicates from dataset specifications - [ ] Test LLMs to construct task criteria based on provided predicates and task description in natural language...