The takes the aggregation to dictionary that was in Pandas and puts it into SQL, which gives improved speed.
The longest missing words (e.g. Psalms, with 29 baseline_ids) now takes around 40 seconds instead of 100 seconds when running on my local server. (Both should be a bit quicker on the deployed API).
The takes the aggregation to dictionary that was in Pandas and puts it into SQL, which gives improved speed.
The longest missing words (e.g. Psalms, with 29 baseline_ids) now takes around 40 seconds instead of 100 seconds when running on my local server. (Both should be a bit quicker on the deployed API).