What is Vertex AI?
Vložit
- čas přidán 9. 07. 2024
- Introduction to Vertex AI → goo.gle/3r428tg
Vertex AI is Google Cloud’s end-to-end ML platform for data scientists and ML engineers to accelerate ML experimentation and deployment. The platform unifies Google Cloud’s existing ML offerings into a single environment for efficiently building and managing the lifecycle of ML projects. In this video, we show how this toolset supports your entire ML workflow from data management all the way to predictions. Watch to learn how you can use Vertex AI for your ML projects!
Chapters:
0:00 - Intro
0:56 - Basic ML Workflow
1:29 - How Vertex AI offers tools for different steps of ML workflow
3:39 - Demo Vertex AI Dashboard
6:37 - Summary
Subscribe to Google Cloud Tech → goo.gle/GoogleCloudTech
Watch more episodes of AI Simplified → goo.gle/AISimplified
#AISimplified
Product: Vertex AI; fullname: Priyanka Vergadia; - Věda a technologie
Another fantastic presentation by Pryianka! Thanks for the hard work putting this together!
The best teacher and coach eva! Thanks for sharing!
Very nice! Thanks for the information as it was really useful. Looking forward to learning more. Hopefully, upcoming videos will include examples of image segmentation, recognition, tables, etc.
Yes, they will!!
Very well explained, thanks Priyanka!
Priyanka is just AMAZING!!!
Awesome!! Thanks the introduction. Very easy in this way you show us.
Thank you!
Congratulations for the fantastic release !
Thanks Hasan!
This is really exciting!
explained simply and clearly, thanks for the clear explanation
Cool. This was really helpful.
Great Presentation. Thanks a lot.
Thank 😊 for valuable information, priyanka..👌
Perfect approach, congratulations Priyanka!
Thanks for your information
This was really helpful
You are an awesome instructor!
Thank you so much Yodi!
I’m excited to hop in and text automl out!
What is the board are you using to demonstrate the demo
Very Nice and informative
awesome explanation
@priyanka - any pointers on how to build a custom model for OCR use case? Better if I can customize one of the Document AI solutions
Thanks a lot: very insigthful
It will help to learn AI ......... I think so ......... Please continue.........
verry well explained.
please take a real-time example (predicting the results and create a pipeline from scratch)
where to collect the sample datasets
Good morning Pryianka, I have been following you for a little while. Here are some questions on my side.
Could you please present one or more of the following:
1- one simple example of an end to end flow in one single episode
2-Whether or not this technology can provide Low Code continuous learning capabilities
3-How it compares to the similar offering of Amazon and Microsoft and why i should focus on this
4-Your plans if any to make more NoCode A.I. appliances available.
All great points - 1,2 and 3 are coming in the next episodes. Follow the series. I have made a note of 3
Awesome Presentation & Great Talent. Yaarunga Neenga? 🙂
Thanks
Good day Pryianka!
Thx for nice and clear videos about Vertex AI!
May I ask one question please?
Suppose we have a deployed data pipeline in Google Vertex AI. Is it possible to make two different accounts to access this pipeline? One account for the developer of this pipeline with full access. The second one is for the analyst's client application, who can only connect to the pipeline via API, pass new data as input of the model (raw data table X) and then get result of the model (Y column).
Best regards, Nikita
Great Presentation, Did anyone noticed she was writing reverse on glass from her perspective? 🤔
Yeah.
She recorded normally. The video has been flipped horizontally. The logo on her shirt is previously prepared.
Yes, like Luciano said, it is post production magic. But I hate to admit it because it sounds so cool as my superpower to write backwards :)
That's how it should work. It is reverse for you not for her
yeah i am wondering, how ?
How can we find out, which model was used in the auto ml by vertex ai that was used for our data training ?
Great explanation!
Thanks!
Nice overview
Cool idea with writing on the glass. Looks like you're able to write in mirror reflection ;)
Hi, is it possible to create Deep Reinforcement Learning models inside vertex?
Nice.
Can I use this tech to train our data from text document into decision making?
Gratious
Very nice
Hi, is it possible to use AlphaGo or Reinforcement Learning process in Vertex Ai?
Thanks for explination PriyankaI
Totally cleared that up fior me. -_-
Just starting, thanks Google!
I used Google AI products and love the easy UX
I really appriciate the lengths that google went to making this video, They have even made an mirroed version of ther logo on the shirt in order for it to look normal when mirroring the video
Hi are there any Dialogflow CX certification training available?
Which model does Vertex AI Auto ML use for image classification task (Ultrasound images) for training?
I love your enthusiasm. Just starting with AI/ML today. I would start with what neurons are and how "stove" and "touch" are mixed to result in "NO". I understand that Vertex AI is about training neural networks, right?
Neurons and etc are not related to this video at all. VertexAI is managed platform for training and deployment.
is there a newer version of this playlist?
How powerful is time series tabular AutoML here compared to offerings by other vendors?
Great
hi, 2 weeks since I asked the next question in the community, can sone one help me? Hi, is it possible to use AlphaGo or Reinforcement Learning process in Vertex Ai?
Please share link to the detailed step-by-step approach of using ML workflow using Vertex AI once it is available.Thanks for the introduction.
All docs & tutorials are cloud.google.com/vertex-ai/docs
I know I should be paying attention to what's being said but is she writing on a clear board *backwards*?! The amount of skill required to do that is boggling! 👏
Idk why no one else is talking about this lol she can just casually write backwards
She write the normal way and after you mirror verticaly the picture
Hello, I have a question. If I am working with a team, is it one person who is supposed to have a vertexxAI account that will be charged? Or will each user, though sharing the same project, be charged in their account?
I have trained (using AutomML) a model and I'm able to fetch predictions from the endpoint. However, when I try to use the model in the following code:
python
Copy code
model = TextGenerationModel.from_pretrained("*********")
I'm getting an error that says:
"NotFound: 404 Publisher Model publishers/google/models/******** is not found."
Could you please tell me how I can use my trained model in this code?
Is there any additional context or information you can provide to help with this issue?
were you writing mirror in this video?
So I think the "data prep" part is actually Trifacta under the google cloud hood.
Is this similar to Azure ML / AWS ML..?
I have some data in a bigquery table and I would like to further manipulate this table using some python code. Will the pipeline outlined in this video allow me to do this before training a model? My understanding is that this could be done in the container section. Any insight on this area would be greatly appreciated. Thanks :)
What I could understand is you can only put code which is used to Train your dataset into the container but if you want to bring the data then you can use the create data set tab itself in which there is an option to add custom data table
Love this episode! What do you use to write on the screen/board?
I am confused! how's she writing inverted english so fluently 😅
Some Google tech
I was wondering the same thing. I'm thinking that the video is inverted. She probably has a an inverted logo on her shirt and is actually right handed. It looks like she has a wedding band on her right hand, which could be another clue that the video is inverted.
I confirmed on LinkedIn that she wears her wedding band on her left hand, while it is on her right in this video. Inverted video. Clever!
@@davidmoreau7589 Nice observation! Yes, this is post production magic :) I am not writing backwards...so hard to admit because I want that to be my superpower :)
Epic
Hi Priyanka, thanks for sharing your knowledge. we are working on the healthcare domain for diabetes patients' treatment requirements. The use case is to containerization and predict the dataset. Are there any provisional steps for K8s and AutoML?
For my understanding of the difference with AI Platform Unified, I need to know what are the difference, especially because a cost raise is planned and what are the addition to justify this new cost rate? What I will earn with Vertex compared to the previous version?
Great question. Does this doc help? cloud.google.com/vertex-ai/docs/start/ai-platform-users
@@pvergadia Wonderful!! You rock Priyanka!!!
this is an AI persona or she is the first human being to be able to freely write back-wards ! 😜great product btw. Looking to learn more to see if this is a good fit for my business.
I get the therotical thing, but can you give us an example from beginning to the end...
I have made Geometrifying Trigonometry theory with whole dataset upto 6th power of all trigonometry ratios geometry
You know that as per Geometrifying Trigonometry,every multiplication of numbers generate 4 different new geometry forms so for every 6th power of each of cos,sin,tan etc there are 4^6 configs which is a large possibility set
For every single value of theta which i have populated from the zero to 360 with 1 degrees step iterations
Now i want to cross verification of each line segment with others to search lots of theorems
These new theorems are (i strongly predict) will become as deep and pervasive as Pythagoras theorem
How the hack are you writing from behind the glass in mirror image so well, so easily? :)
can you still deploy models for edge devices?
Yes, you can. Here is the documentation and the video is coming later in this series. cloud.google.com/vertex-ai/docs/training/automl-edge-console
and she mirror-writes on the go!? ...amazing
Do product managers at big tech all need to learn how to write backwards? She seems to be really good at it.
You need to create a troubleshooting session
how is this more beneficial than just doing everything within a jupyter notebook without vertex AI?
I would like to have the opportunity to develope and put to use my world of photography and business equity markets to the new world of learning. This may be the way?
👍
I love how she is presenting and writing the words in reverse at the same time. Must require a lot of training data to get right.
Presentation IBM style format.
This is not the end-to-end ML lifecycle. Online model performance monitoring is missing.
This is all great but can you at least clarify what's the difference between this and your now old(?) AI platform? This is so confusing!
I think they've unified a few products (incl AI Platform) under Vertex AI. cloud.google.com/ai-platform/ redirects to Vertex AI page.
Google: 1000000000000000 interviews of how to invert a binary tree
Does google make t shirts with a reverse cloud logo on them so it matches when they flip the video?
honestly I'm a bit lost with all gcp AI products. It's a mess
Agree with you, the products are certainly awesome and way better than anything that AWS has, but they are confusing everyone with so many products that seem to do the same thing and not properly explaining the difference.
She's actually a robot. Who can write backwards so well?
Google
Cloud
Machine
Developer
When you hit that Uno reverse card.
How is she writing backwards so well
Oh they flip it
I would like to sigh up?
How does she write in the opposite direction?!?! plop!
She writes on a transparent glass in reverse but the final video is flipped horizontally to show us in proper way.
नमस्ते मैडम
अश्लील वीडियो से बच्चो को सुरक्षित रखा जा सके ऐसा आइडिया है मेरे पास
किसी भी चैट एप्लीकेशन के लिए जो अभी पूरी दुनिया में नहीं प्लीज मैडम कांटेक्ट करे
wow did you have to learn how to write backwards for these videos?
I dont need your AI IN GERMANY