Develop the table classifier. It should read the sample vectors as a resource so that they can be read at startup of the pipeline, and then it should tag each table with a cluster id.
The table classification should only run when the vectors resource is available, so that if no vectors are available, the tables don't get classified.
Develop the table classifier. It should read the sample vectors as a resource so that they can be read at startup of the pipeline, and then it should tag each table with a cluster id.
The table classification should only run when the vectors resource is available, so that if no vectors are available, the tables don't get classified.