Mapping Click-ML to MLOps

Now I have Click-ML to automate ML flow - data cleansing, training, parameter optimization, prediction, insight extraction and reporting. Other people are talking about MLOps. Since my app is working just fine - am I missing something?

The most important thing - both are not open ended. You can improve your data and optimize your model, so this part of the ops is continuous and living.

MLOps emphasizes Design versus Development (others being Operations), whereas in Click-ML I focus on Insight and Action. I always ask the question 'How do data help me make informed decision?'

In conclusion, I don't have any serious thing to worry. I need to remind myself 'How do cloud services help me scale that best serve my clients?'.

(Photo from Ml-ops.org)

#datajourney