Machine Learning Modeling Pipelines in Production

I have completed another course in the Deep Learning AI specialization "Machine Learning Operations", namely "Machine Learning Modeling Pipelines in Production" the course is available on Coursera here. There is a certificate issued after completing the course, an example is here.

The course is advanced and covers a number of very interesting and important topics, such as:

1. NAS (Neural Architecture Search).
2. Model resource management techniques of dimensionality reduction, including PCA (Principal Component Analysis) and NMF (Non-Negative Matrix Factorization) and quantization and pruning.
3. High-performance modeling and knowledge distillation.
4. Model analysis and model performance analysis.
5. Interpretability.

From my point of view, some topics are described very well at the right level of depth and providing sufficient additional information and references to the publications.

However, some other topics would need additional research to understand all the concepts and principles, such as knowledge distillation or interpretability.

Overall, I would recommend the course since it covers very important aspects of the ML, however I would book more hours than described (23 hours) to read more about topics such as knowledge distillation.