A training project, or a POC project, is the term used for a project carried out before one or more production projects. It is often smaller in size or duration, and its objective is to lay the groundwork before the production projects. This type of project allows for upskilling labelers on the tasks, as well as properly configuring the labeling tool, applying the correct labeling process, and resolving edge cases before starting production.
At People for AI, we refer to a project as a training project once it has already been paid for. It is worth noting that the labeled data during the training projects may be of lower quality. The data scientist is free to use them or not to train their model.
Training projects and POC projects are synonymous. Therefore, going through a training method brings numerous advantages (see POC Project).
Synonyms: POC Project
