Implementation of Machine Learning Based Systems into Production

This paper provides a brief overview of the main stages associated with the introduction machine learning based systems into production. The relevance of this problem is shown and, based on the analysis of the experience of the IT industry, the main features of each of the stages are highlighted. As a result, it is shown that the total costs of embedding the system into production, taking into account all stages above, can significantly exceed the costs of creating and training the model.