Agile SD approach employs DevOps strategy for the speedy delivery of products. ML processes lifecycle, due to its ‘self-learning’ capabilities, remains unattended. ML-Ops emphasises on agile development in ML covering stages like data acquisition, cleansing, model development etc. An active feedback mechanism monitors performance (precision, recall, accuracy) and alerts ML-Ops teams on the need to update the models when there is a data or concept drift.

Click here to read the full story