Chevron Left
Back to Deploying Machine Learning Models in Production

Learner Reviews & Feedback for Deploying Machine Learning Models in Production by DeepLearning.AI

4.5
stars
318 ratings

About the Course

In the fourth course of Machine Learning Engineering for Production Specialization, you will learn how to deploy ML models and make them available to end-users. You will build scalable and reliable hardware infrastructure to deliver inference requests both in real-time and batch depending on the use case. You will also implement workflow automation and progressive delivery that complies with current MLOps practices to keep your production system running. Additionally, you will continuously monitor your system to detect model decay, remediate performance drops, and avoid system failures so it can continuously operate at all times. 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: Model Serving Introduction Week 2: Model Serving Patterns and Infrastructures Week 3: Model Management and Delivery Week 4: Model Monitoring and Logging...

Top reviews

MN

Apr 21, 2022

This course is essential for data scientist if they want to embark on the journey of data scientist in industry. I learned a lot of useful techniques. Thank you team!

RF

Sep 19, 2022

Great course with tons of meaningful information and excellent hands-on material. Also videos and lectures and well designed and very well explained

Filter by:

1 - 25 of 50 Reviews for Deploying Machine Learning Models in Production

By Enrique C

Mar 14, 2022

Has some good and useful content but like the rest of the courses in this specialization it looks a lot like a Google cloud infomercial. The graded labs are ok. I have mixed feelings about ungraded ones as a few are really good and some others are a waste of time. I think that students need to be harder in the way they rate these types of courses to force the vendor to deliver quality labs end-to-end.

By Roger S P M

Oct 2, 2021

Robert's lectures are terribly boring and there was no work to make his slides useful, they are just the words he is going to say.

By Stefan L

Feb 28, 2022

If you are doing the entire MLOps specialization, this coures won't bring much insight. If you don't you might learn something, i.e. regarding model serving. Unfortunately the labs are pure copy/paste exercise (qwiklabs) and do not yield any practical inisght. A missed opportunity.

By Jordi W

Sep 30, 2021

So you have a fairly good understanding of ML modelling techniques, you played around with code in Jupyter notebooks and perhaps even got a TensorFlow docker image with GPU support to run on your local machine. You readily admit that there always is more to learn about modelling techniques, but you wonder how models run and are made available to users in a production environment? This course/specialization dives into just that question and a wide set of related subjects. A most important dimension of ML.

By burhan r h

Jan 12, 2022

I was hoping for a final project that I can use in my portfolio because the course content is so much and not easy to digest

By Akie T

Apr 22, 2022

Good overview of major concept in the field, but expect to get just conceptual ideas and long to-do list of what you need to study somewhere else. Exercise (both graded and graded) are buggy and wasted a lot of time on non-essential details (like setting up the environment or just trying something in a different PC).

By Kevin

Jul 16, 2022

This course is exceptional. Why? Because even for people that have worked in production environments (which the course is geared towards) professionals often ask themselves if there are tools they're missing or not optimally taking advantage of when considering new production grade ML models. This course provides that path of "best practices" when it seems like most cloud-based courses are 1-off's of a specific tool but not how they are, or can be, integrated together. Coming from a much heavier AWS-based knowledge it was additionally refreshing to get up-to-speed with what GCP is offering. The education around getting up with Kubernetes and Kubeflow was great. Often it feels like productionalizing ML models is hacking together components to get any solution rather "knowing" a best path. Again, this course does a great job of setting some finite path with different tools (albeit production machine learning is fairly subjective based on company requirements/budgets/etc..). I feel confident I now know enough about tools inside of GCP to help make those artistic decisions about when and why I might opt for more production-grade ETL tools (Dataflow + Apache Beam) and when an "easier" batch processing setup with less complexity is merited. Learning basics of specific tools like Kubernetes was also a big plus.

By M O

Jun 7, 2023

I have to choose from 1 star, but my feeling is that the number of stars is -100.

Due to a stupid and crazy bug in GCP's Lab, I couldn't complete it and it was extended for a month.

I want you to stop using GCP.

Lab support is too bad. I'm not taking Coursera to waste time and money.

I made a complaint to the GCP Lab, but the mentor from Coursera only gave me a guide to request help from the GCP Lab.

After all, courses that collaborated with GCP or vendors are trash.

Please also reflect on Coursera. Educate GCP. And please sincerely apologize to me for the trouble I caused.

By James B

Jul 5, 2023

Seems like an advert for GCP and the GCP based practicals are just copy-paste exercises that aren't useful for learning

By Michalis M

Oct 31, 2023

Very interesting course, covering a lot of stuff horizontally in the landscape of ML-Ops. As others have pointed out as well, it provides the spark to exploring further by yourself, which is needed to truly master the concepts provided. If I had to comment on further improvements, I would suggest, to leave out some code and commands to be implemented by the student. I get that it will make the course much harder, but it will "force" people to really dig into stuff and search the topics, rather than passively execute the provided content. Nevertheless, it truly covers a lot if interesting areas.

By Arthur F

Oct 2, 2021

pretty helpful broad overview of some of the tools and techniques used in deployment of ML models. Gives a good starting point for personal implementation since the field is clearly deep and fast evolving

By Rubén Á F

Sep 20, 2022

Great course with tons of meaningful information and excellent hands-on material. Also videos and lectures and well designed and very well explained

By Travis H

Dec 19, 2021

Very insightful, with a good high-level explanation of challenges surrounding model usage and deployments in a production environment.

By Atul P

May 29, 2022

This is really a good learning with real word prodtion deployment. There are many things which we got in this learning.

By Eoin B

Feb 6, 2022

Really enjoyed it however to get he most out of it, the time commitment is large

By Hieu T D

Aug 3, 2023

good course !!!

By Alexander N

Jun 6, 2023

I would rate this course as satisfactory/good. I achieved my learning goal and appreciate the effort of authors and tutors in preparing the course. Things to improve: - learning topics sequence overall - some of them are repeated or feels like off the place - the material presented in videos are quite often just 'reading the slides', scarce of examples and intuition explanation - I would love to see in the very early part of the course multiple possible architecture options for ML model management, serving and monitoring and their correlation with automation/maturity tiers. Thus it makes it clearer why we investigate deeper one or another tool. Otherwise I felt a bit out of context doing first 3 modiles of the course.

By Stefan S

Jan 12, 2023

I especially liked the ungraded labs where tools like Docker and Kubernetes are explained step by step. The Qwiklabs were a bit harder to follow sometimes.

I would recommend introducing labs where the student has to do more him/herself. This way, you can really understand the concepts better. With the Qwiklabs I sometimes simply inserted the commands but didn't really understand what I was doing

By John T

Dec 3, 2023

This course was less heavy on the math - more on process - I have some idea of the processes I will have to put in place to use these products and techniques where I work. But I am not confident I understand it well enough to set it up and run it.

By Cristian J D A

Jan 6, 2023

Great specialization but to much information about google and its platform, would be nice to lear about diffetent platforms

By AG S

Apr 21, 2023

It would be better if the Google Labs were not graded. The labs keep crashing and are very slow to run.

By Sritam S

Jun 17, 2023

this is not working the penguin file is not forming on cloud all ways showing error.

By Sagar D

Jun 27, 2022

difficult to understand

By Juan P J A

May 15, 2023

I really recommend this course as a general introduction to MLOps and for those interested in the tools used by AI teams (data scientist, ML engineers) for rapid, optimal and secure deployment. Given the wide coverage, with great supplementary reading, and very guided labs, it would require much extra time to deep into the different subjects and recommended tool.

By Gordon L W C

Oct 12, 2021

This course is what I think is missing in the market. A machine learning course with much emphasis on the practical aspects of running a machine learning platforms. I recommend it to anyone who is looking for the next step after you have finished training your model in Jupyter notebook. It is not the end but only the beginning.