Overall a good curated course to help understand the GCP offerings and high level architecture of how their offerings fit in the current landscape. Easy to follow along as this was fundamental course.
This course really helped me in understanding exactly 'How the Big data and Machine learning can be used in Cloud' and 'The ease to use it'. Thank you for summing all the fundamentals in this course.
교육 기관: Brett W•
I find the course material (videos) don't actually cover the questions asked in the module review. Besides that, the labs are incredibly slow - and seem a bit buggy.
교육 기관: Jake•
I enjoyed the course and the way labs are done is excellent. 3 of the labs were just spinning up and running the same notebook, I think this was a lost opportunity.
교육 기관: Gerardo N•
The course is great, but some of the labs are not well maintained, which can be very frustrating.
교육 기관: Mohan K•
The environment is not very stable and performance is not very great
교육 기관: Linda L•
The labs were horrible. I never knew if they would run or not.
교육 기관: Kushal J•
Great content, let down by technical issues using Qwiklabs
교육 기관: Hanumantha R V•
Very Basic course.. But helpful to get started with GCP
교육 기관: dpeif d•
Some labs are very repetitive and feel useless
교육 기관: Sandhya P•
Labs have a lot of errors with grading
교육 기관: Tamás-Marosi P•
Just from the lessons the quizes questions were hardly solvable. Of course with the review video, where the answers were publicated, reaching the 100% was challangless.
The Labs were correct, but there were some bugst and errors, when the labs steps say something but with these steps the environment throw errors. Again there were videos where the presenter shows the steps, which were more or less the same as the labs steps, but with these plus steps the environment worked without errors.
In the labs steps there were fill out type questions, but I wasn't able to fill them out. It would be great if these questions will be implemented into the quizes.
교육 기관: Celia M•
There is valuable information in the course and some good exercises, but they are hidden among many information which is not so useful. In overall it is an entry class to use the technology, but many customers already know this sort of things because they tested our products or the features we comment here are similar to features of other vendors' products. Their reaction when they are presented the technology for the first time is: "Thank you. My questions and issues are quite specific and not reflected here".
교육 기관: Paul C•
had to contact customer service for qwiklabs twice due to it bugging out & becoming unusable.
Also if you go outside of the "Script" & miss one step in the instructions for an exercise, it becomes really hard to figure out what it was that you did wrong.
Lastly this is definately not a "compressed course". It could be completed it one week if you literally dedicated a whole week of time to it but for those of us who are doing this part time & actually want to absorb the content, expect to plan for 2-3 weeks
교육 기관: Lourdes R•
Steps missing in the labs. The documentation cannot be downloaded from just one click. The instructions in the lab miss some steps. It would help to have a diagram with the steps, incomes outcomes and things to consider to avoid errors.
교육 기관: Vinod K•
The contents covered are good. However, the labs had many issues. I spent more time debugging and talking to tech support than I spent doing the lab exercises.
Overall, I feel not the best utilization of my time.
교육 기관: Jón A T•
Superficial. Little learning occurred. Basically a product presentation.
교육 기관: jiawei•
many labs doesn't work
교육 기관: Felix E•
TL;DR: I sadly have to say that this course was absolutely not worth the time. I would recommend anyone looking to get started with Google Cloud to just go through the examples in the Google Code Labs, since you'll get about the same information in a much smaller timeframe.
I can't recommend this course to anyone for the following reasons:
(1) For the most part, it feels more like a Google self promotion than an educational class.
(2) Most of the content in this course is based on the exercises from Google Codelabs (or rather, they link you to the exercise on the google platform), which themselves are all right. The video lectures, however, rarely add anything of value, definitely not with respect to the amount of time they take to complete. Overall, a rather low "production value"/"-quality" compared to other courses in Coursera.
(3) The course attempts to showcase how easy some complicated tasks are in Google Cloud Engine. However, you're just given the task to execute long and rather complicated scripts that Google has provided just for this course. Those scripts and what happens in the background is not explained in depth, instead it's often emphasized how easy everything is with GCE. No mentions of what to do when you DON'T have a huge amount of code already provided by Google.
(4) The course content is not adapted well for the Coursera format. Around 50 chapters, partly only with 15 second videos, so you have to scroll a lot and it's hard to see how content was intended to be grouped. The tests/mutliple choice exams are also very badly done, feels like they were made with the lowest amount of effort possible.
교육 기관: Ekaterina P•
Very raw. Have to restart labs several times, because sometimes they just do not work. Just that (like there should be a button(exists in video), but you can't find it). Very frustrating. Not to mention that for your own money they are advertising google-cloud to you... :)
Google cloud itself is buggy for now...
On technical side of the course: it is made to maximize the number of sections and videos. Just like some ... code that maximizes lines of code. Yes, it's like 10 second videos in the separate section. Many of them. I like short 3-10 min videos, but 10 second ones? Labs that officially take 1h, but if google cloud is not bugging it is 15-30 min. Labs where instructions tell you to stop half way through, because the rest would be in the next lab!!! (so you would have to redo those steps, yeah). Probably maximizing the number of labs too... Lots of frustration!
교육 기관: Deleted A•
This course doesn't introduce you to the concepts; more so it is an advertisement for Google Cloud Products. The labs don't explain how things work, they are just naive click-along activities.
교육 기관: Zhe S•
Very bad experience with labs facilited by Qwiklabs. The scored cannot be saved by Qwiklabs correctly. I have to retried several times and upload the screenshot to manually update my score!
교육 기관: Sainath R•
Some lab exercises have issue and the score doesn't get updated even though the all steps are completed
교육 기관: Ivan R•
most options don't match with current menu
교육 기관: hfculver•
Very shallow course content.
교육 기관: wenhui z•
The lab is outdated
교육 기관: Nikhil M•
In this course, I learned How to use the Google cloud platform(GCP) and it's tools like BigQuery, Cloud Storage, Vision, Dataproc, Pub/Sub, Dataflow, compute engine, etc.
In GCP, we can generate the instance of Virtual Machine(VM). It's a serverless platform (Google has it's own data centers). We can develop a complete software through GCP.
IN GCP, we can build custom models. It is very handy to operate for BigData. The Data in GB, TB, or PB can be processed in seconds or minutes on GCP.
Also, I deployed the ML model for Classifying Images with Pre-built ML Models using Cloud Vision API and AutoML.
In this model, we classified the Images of clouds in three categories., viz cirrus, cumulonimbus, and cumulus.
The cool features of AutoML and Vision API-
-We don't have to write code for building the machine learning(ML) model.
-AutoML decides the dataset splits for training and testing.
-If you are working with a dataset that isn't already labeled, AutoML Vision provides an in-house human labeling service.
- We have to just evaluate the model by adjusting the Confidence threshold and the confusion matrix.
Sometimes the training time will be more because of large datasets, node training time as well as infrastructure set up and tear down.
Though it is cost-efficient because you have to pay for the memory you use, The time processing takes place(for Training the nodes in ML), etc.
The bottom line is GCP offers IaaS (Infrastructure as a Service) in the form of Google Compute Engine, and it offers Paas (Platform as a Service) in the form of Google App Engine. As for FaaS (Function as a Service), GCP offers it in the form of Google Cloud Functions.