Label Generation Process

Once you've uploaded data and defined variables for your project, it's time to generate labels.

Each generation takes time and resources. To make the best use of them, we recommend a process like the following:

1. Generate for one patient to check data and variable setup.
   1. Add or remove variables if you'd like
   2. Change variable definitions and instructions, optimizing instructions if you haven't yet
2. Repeat single patient generation as necessary until you’re satisfied.
3. Generate for a portion of your dataset. 
4. Review that portion in Label Review, making data edits as necessary. These edits will automatically be included as training examples for the next generation.
   1. If you are editing or removing a lot of labels for a specific variable, consider modifying or optimizing the instruction to get better results.
5. Repeat steps 2 and 3 until you’ve reviewed all of your data.

Once you’ve reviewed a data point, Brim will not overwrite that data point when you do a new generation unless you explicitly instruct it to overwrite human-reviewed labels.