When you state at 1:50 that most AWS customers use a VPC to secure their services, you were talking about Lambda at the time. the VPC isnt featured in the arch diagram. could you be more specific about the way the resources inside the VPC are configured? Are we talking about relatively static data that does not come in the restful api request that also is needed to render an inference by the model endpoint?
Or do you mean that the VPC is placed around the ec2 instances that the endpoint uses to calculate the inference, at left, inside the auto-scale and load-balanced multi-AZ environment? and as a result, the Sagemaker ML endpoint instances used for deployment are NOT fully managed the way that the Sagemaker notebooks ARE fully managed, is that correct?
YES. I mean, "NO!". Whatever you do, DO NOT do it! Doing anything is the wrong thing to do. The police are watching, and your phone is now a tool of foreign agents. CAVEAT EMPTOR! Long live the Revolution! #ExplosivePhone #Russia #UFO's
If confusing people was the aim of the video, then Congrats, u did it !
5 years ago, but it is still valuable, great job Emily!
in simple and easy words you explained lot of thing.
Thank you Emiliy 💯
Is there any tutorial on Docker image based deployment ? This video does not provide enough information.
Hey Harshit! did you find any tutorial or solution for this?
Sarvesh Thakur not yet.
Sarvesh Thakur not yet.
Sarvesh Thakur not yet.
Can you upload the notebook? I would like to review some of the basic concepts related to SageMaker.
Regards.
Emily Weber is a sagemaker boss: 99.99%
Thanks Emily !!. Fantastic Tutorial.
Thank you! 😀
Great video!! How can we turn off our real-time endpoints when not in use with Lambda?
When you state at 1:50 that most AWS customers use a VPC to secure their services, you were talking about Lambda at the time. the VPC isnt featured in the arch diagram. could you be more specific about the way the resources inside the VPC are configured? Are we talking about relatively static data that does not come in the restful api request that also is needed to render an inference by the model endpoint?
Or do you mean that the VPC is placed around the ec2 instances that the endpoint uses to calculate the inference, at left, inside the auto-scale and load-balanced multi-AZ environment? and as a result, the Sagemaker ML endpoint instances used for deployment are NOT fully managed the way that the Sagemaker notebooks ARE fully managed, is that correct?
Is there any tutorial on turn endpoint off when not in use with lambda ? This video does provide zero information.
Very informative. Thank you for sharing!
Can SageMaker endpoints support RCP endpoints?
Excellent video, thanks for the thoughtful explanations!
how can I download the input data from Amazon S3 and get it analysed by a model i have already deployed on sage maker already.
Were you able to figure this out?
Thanks for the video!
why AWS tutorials are so confusing? Why can't they show a simple real world application being deployed end to end.
because it's all sales gimmicks from people who can't actually develop and like to crutch on presentations.
What a sticker on the mac.. can't concentrate.
Um...Whaaaat? I found this on my phone and my phone is running out of room and is telling me to delete these. Will my phone blow up if I do?
YES. I mean, "NO!". Whatever you do, DO NOT do it! Doing anything is the wrong thing to do. The police are watching, and your phone is now a tool of foreign agents. CAVEAT EMPTOR! Long live the Revolution! #ExplosivePhone #Russia #UFO's
This is probably the worst introductory video of any AWS product.
You are wrong, this is not probably the worst, this is definitely the worst !