Bringing a machine learning model into the real world involves a lot more than just modeling. This Specialization will teach you how to navigate various deployment scenarios and use data more effectively to train your model.
In this third course, you will: - Perform streamlined ETL tasks using TensorFlow Data Services - Load different datasets and custom feature vectors using TensorFlow Hub and TensorFlow Data Services APIs - Create and use pre-built pipelines for generating highly reproducible I/O pipelines for any dataset - Optimize data pipelines that become a bottleneck in the training process - Publish your own datasets to the TensorFlow Hub library and share standardized data with researchers and developers around the world This Specialization builds upon our TensorFlow in Practice Specialization. If you are new to TensorFlow, we recommend that you take the TensorFlow in Practice Specialization first. To develop a deeper, foundational understanding of how neural networks work, we recommend that you take the Deep Learning Specialization.
SZ
I learned a lot from this course about how to optimize TensorFlow data pipelines and how to create public datasets. Thank you! - Steve
SS
Just what i needed. Would have gotten 5 starts, if they fix the last lab.
RR
First 3 weeks are really nice but for me week 4 was a bit tough with very less explanation
YO
Debugging exercises due to errors in indentation sounded stupid in the first place. But the joy of finally getting a "yes" in the assignment auto-grader beats them all.
VM
Was not that much engaging because the lectures were not linked properly and were lacking examples to support the content
SB
I was really looking forward to learning efficient data pipelines and that is utterly what I learned here in this course. Best in its class.
FF
T​his course many for challengging in code but, before it, i learn it in my college. So, i do this assignment.
NB
This seemed very helpful and hands on. I can't wait to try this on my own.
EJ
Just completing this course and I have already applied the knowledge gained to solve a problem.This course is quite practical.
PB
covered lot of scenarios that helped to understand the options out there
RN
Had a good time learning about tensorflow datasets. Probably the last assignment was too taxing to complete. It was fun overall.
RV
Excellent content, but the design of Exercise 4 tainted the experience somewhat by the end.
Showing: 20 of 108
I'm sorry but, it does not seem realistic pipelines, it clearly show the capability of tensorflow, but real world data pipeline on my point of view is completely different from that. I was expecting something like how to handle large amounts of data coming into the cloud, or onpremise cluster, and get it into a retraining pipeline, improving the models... but was completely different... If you are expecting something like, How to retrain a large model with large amounts of new data, realtime... that is not the course for you.
I love Andrew and Lawrence, but this last specialization is not at the same level from the other 3 from Deeplearning.ai, you guys should consider rethinking it using more Cloud deployment strategies with Tensorflow, like delivering APIs that requests model inference, and retrain automatically, using Google Cloud, Sagemaker, Azure whatever..., integrate it into a MLOps/DevOps model, and delivery at scale, at edge, that is the real world of deployment in my view...
People have complained about the last assignment's compilation problems for months and it still has not been solved. No teacher answers students' questions in the forums either, so prepare yourself for spending hours reading the forums for the last assignment and resubmitting it till, without learning anything new, you realise there was an extra "s" in the name of a variable during the videos and that was causing the compilation problem. Interesting course nevertheless !!
Some issues with notebooks. This is still in beta. Absolutely no help with the technical setup (notebooks and the Tensorflow datasets). Needs to be debug a couple of times..
Unfortunately this course is extremely weak. Tons of poorly explained code and nothing else
Every nice part of this course is obscured by the shamely developed exercise in week 4. It's a GIGANTIC pain is the ass, as it was badly written and the debugger does not help.
I decided to quit due to the malformed exercise in the last week. Visiting the Forum I found out that many students had problem with this exercise.
In my oppinion, it's a waste of time fighting with it.
The course materials were a detail explanation of the data pipelines in TensorFlow. It could have been better by explaining the materials slowly and with multiple simple assignments and then finally a mega assignment.
The last exercise does not seem complete. There is too less help about solving the excercise - moderators do not help.
Really important topics if you want to operationalize your ML models. Final exercise was pretty hard to debug to satisfy automatic grader which produces a lot of frustrations for learners. Should be redesigned.
Basically, i liked the course as I got lots of new knowledge about data pipelining using tensorflow. However, the one downside of the course is the last week's assignment, which is just awful.
I was really disapointed by this course. Especially the last exercice on imbd faces I absolutely did not understand anything, just copied some code from the VIdeo ....
Laurence cares deeply about the students. Not only about what they learn, but that they actually enjoy and learn it. What a fantastic teacher.
Excellent course both for Data Scientists and Machine Learning Engineers!
Last week is a total disaster
While this is meant to be a "get your feet wet" specialization that has you use basic TensorFlow APIs to do things. this particular course is below the bar even for that. Exercises consists of cut-n-pasting large swats of code and asking meaningless questions about spelling of API symbols (quick, which one is it: word_order_isn't_learning, word_isn't_order_learning, learning_word_order_isn't, or order_isn't_learning_word?). The course ends with a seriously frustrating cut-n-paste exercise where you get tested on important topics like finding the right sentence to copy from a wiki.
Worst Course Ever !
I understand why most of the students are furious about, but content wise, it one of those extremely helpful and important courses in Coursera. Really loved it!
so tired
The last graded exercices is like black magic! After the first run the notebook do not run anymore. You have to do everything locally, and submit without using the online version, and of course, hope for the best. Most of the code to add is very easy to find, but making it work is really a challenge. Some topics went to fast, some others were too long. Overall still correct.
This course is quite essential, yet it is prepared too quickly? I had a hard time on Week 3 and Week 4 because the exercises are hard to grasp.