Chevron Left
Back to Machine Learning Modeling Pipelines in Production

Learner Reviews & Feedback for Machine Learning Modeling Pipelines in Production by DeepLearning.AI

4.4
stars
408 ratings

About the Course

In the third course of Machine Learning Engineering for Production Specialization, you will build models for different serving environments; implement tools and techniques to effectively manage your modeling resources and best serve offline and online inference requests; and use analytics tools and performance metrics to address model fairness, explainability issues, and mitigate bottlenecks. Understanding machine learning and deep learning concepts is essential, but if you’re looking to build an effective AI career, you need production engineering capabilities as well. Machine learning engineering for production combines the foundational concepts of machine learning with the functional expertise of modern software development and engineering roles to help you develop production-ready skills. Week 1: Neural Architecture Search Week 2: Model Resource Management Techniques Week 3: High-Performance Modeling Week 4: Model Analysis Week 5: Interpretability...

Top reviews

JS

Sep 13, 2021

Excellent content and lectures from Mr. Robert . Thank you very much Sir for the excellent way of explaining these difficult topics . Thank you !!!

MB

Oct 20, 2021

I enjoyed this course a lot. It gave me a lot of ideas on how I can improve my models and make my workflow more efficient. Thank you.

Filter by:

1 - 25 of 81 Reviews for Machine Learning Modeling Pipelines in Production

By Folkert S

Sep 18, 2021

I thought this course was ok. On the one hand, the theory that is taught is quite general and trivial, while on the other hand, the technical focus is mostly on Google's tools and deep learning. As getting machine learning to production is an advanced task and requires a broad set of skills, I would've, for instance, expected this course to be more on structured data. Also, most of the labs, especially the GCP ones, feel just like copying and pasting some commands, it's not that challenging and therefore I didn't learn a lot there.

By Panagiotis S

Jan 25, 2022

Poor content on this course as well. A bit of intermediate machine learning concepts that we all have seen a thousand times and a bit of mlops. The instructor was always always reading only whats in the slide. Graded assignments on GCP were just copy/pasting the code and had no difficulty or needed any critical thinking or skills. Again focused ONLY on Tensorflow libraries that are incompatible with models from other libraries like Pytorch.

By Xavier D

Oct 13, 2021

I find this course extremely hard to follow, some main and tricky concepts are only covered by a mere sentence in the lecture.

By Arturo M

Jun 26, 2022

I'm a big fan of Andrew Ng's ML courses. However, I'm very dissapointed with this one, for several reasons.

First, the instructur is not nearly as engaging as Andrew Ng himself. Most of the time he basically reads through the slides in a monotous way.

Second, the course tries to cover too many concepts. Instead of selecting a few core topics and explaining them in detail, the course seems like an ennumeration of ML concepts. Most of the times, the explanations are way to shallow to be of practical use.

Third, the exercises are quite poor. Most of them are just plain Google Cloud tutorials on Quicklabs.

By Nithiwat S

Jun 29, 2022

Just like the previous course in the specialization by the same instructor. He bascially reads from slides with little to no explanation. He barely explains any concepts, gives examples to help develop understanding. Some concept is unclear and poorly explained. Every single lecture, he just reads from scripts. It's very frustrating to learn. Coursera, in my view, is strong in delivery content that's more technical, more engaging, better explained technical concepts. But this course fails to deliver that. Watching many videos on Youtube is better than learning from this instructor. It is just terrible delivery, and I wish it was Andrew Ng who taught it. The only good part is the labs. Labs are well prepared and help with the study.

By Peter W

Aug 9, 2021

Covers a lot of content at a high level. One slight criticism is that the graded exercises focused on Google cloud and didnt require much thought. The ungraded labs on the other hand were quite interesting.

By Roger S P M

Sep 5, 2021

So Boring!

By Phillip G

Aug 14, 2022

Except for the huge amount of Google TFX sales talk, the biggest disappointment is that the instructor just reads the slides plainly without going over the key concepts in depth. Some of the concepts like PCA, they should be explained much more clearly, but what he does is just showing a rotating image and throwing out a few quizes. I mean, what's the point? I get far more knowledge from an Youtube video than mostly any of the concept explaination here. I have to double my time learning them, because there are rarely any good, down-to-earth explanations. Everything here is super fast, plain reading, and "Google stuff is the best".

By Ashwani K

Aug 7, 2021

Some of the topics were too advanced and instructor assumes that we know those basics. It felt rush through little bit and more of reading slides then explaining at many places

By Stefan L

Feb 12, 2022

While this course conveys lots of interesting and relevant knowledge, it's labs do not. In fact, they are usually just copy/paste tasks based on existing GCP tutorials leveraging qwiklabs.

By Alexander N

May 19, 2023

The course is under whelming. The video materials are mostly about listing different methods with very minimum focus on intuitions. It is not optimal, but okay. Technical state of the labs were horrible when I did the course. I faced issues with every third of them (especially on Google Colab). It was mostly driven by update of Python there, yet, I would have expected Labs to be regularly tested and fixed timely. On the bright side, I learnt how to run configure and run the labs locally. But, sorry, I didn't subscribe for it.

By Hitesh K

Jul 18, 2021

So far the most informative course in this specialization. This course has actually taught me how different is ML in production than doing simple Ml stuff on notebook for academic or research purpose. You get to see the bigger picture, i.e, different and bigger constraints that needs to be addressed for deploying any model to be on systems, specially edge devices.

By Cosmin K

Aug 16, 2021

Great material, insigthful notebooks and a valuable review of numerous concepts and tools! The course set me on track with steps to take and pitfalls to evoid. Thank you! Now is practice and continous learning from my part.

By Kiran K

Jul 22, 2021

Good But More practical needed with theory

By Thành H Đ T

Aug 24, 2021

wow, Its very good

By Hieu D T

Aug 15, 2021

A bit dependent on GCP, took me quite a decent amount of time to do network setting. You should use your own internet, do not use one behind corporate proxy like I did. Materials and guides are great.

By Andrei

Sep 9, 2021

need to improve the explanation of topics

By Karol J

Jan 5, 2023

After 1st course in this specialization, this one is disappointing.

In most cases, slides were containing text only and were more distracting than helpful.

Many important concepts were insufficiently explained with only one or two sentences.

Graded labs were ok, but most of the ungraded labs were more lectures than labs (no code was written - only thing you had to do is run cells in colab)

In addition, it feels like the course focuses only on google environment rather than in general ML concepts.

By James B

Jul 5, 2023

Feels like an advert for GCP, the GCP based practicals are just copy paste exercises

By Will G

May 7, 2023

This particular course is a confidence booster on the intricacies surrounding model building and optimization. Super cool concepts like Knowledge Distillation and papers like "Optimal Brain Damage" excites you on your seat. Thanks to Robert Crowe, the team at TFX and Coursera for this invaluable treasure.

By James A

Dec 7, 2021

This course gives a very good overview of this topic. Very relevant to my day-to-day work (not academic, or too focused on research, etc.), it is well presented with good context and examples. Obviously a lot of hard work went into creating this course. Good learning experience, +1, thanks!

By Jonathan S R P

Sep 28, 2021

I strongly recommend this course to anyone interested in MlOps and how to manage a ML pipeline in production, i learn a lot about pipelines, distillation and interpretable models. Can wait to put all this knowledge in practice :)

By Travis H

Dec 19, 2021

Consistent with the other courses in the specialization for MLOps -- very insightful with good coverage of content that is relevant to the pipelines and automation requirements for proper production support.

By Umberto S

Aug 29, 2021

Great course! One of the most clear and extended courses by DeepLearning.ai. I think It covers in an excellent way all topics to understand what MLOps is and how to approach it in the right way.

By Jitendra S

Sep 14, 2021

Excellent content and lectures from Mr. Robert . Thank you very much Sir for the excellent way of explaining these difficult topics . Thank you !!!