How to Deploy ML Solutions with FastAPI, Docker, & AWS

[@ShawhinTalebi]

More on Full Stack Data Science👇 👉Series Playlist: kzbin.info/aero/PLz-ep5RbHosWmAt-AMK0MBgh3GeSvbCmL 💻Example Code: github.com/ShawhinT/KZbin-Blog/tree/main/full-stack-data-science/ml-engineering

[@waleedashraf9t]

One of the best video available. on the internet for deployment.

[@ShawhinTalebi]

Glad it was helpful :)

[@FREAK-st6kk]

One of the best aspects of AWS Elastic Cloud is how seamlessly everything comes together, whether you're using FastAPI or Docker. It's all integrated beautifully.

[@divyanshtripathi4867]

This is such a great video, no nonsense straight to the point!

[@pawe5560]

Hey Shawn, videos on FastAPI and Docker from you would be great.

[@ShawhinTalebi]

Thanks for the suggestion! I'll add it to my queue :)

[@miraclechijioke1213]

This was so simplified. thank you Shawhin

[@jonathankerr2479]

Fantastic video. Thank you for this!

[@Coret-with-c]

Super detailed explanation, thank you!

[@tusharrohilla7154]

It was amazing ...love this type of content❤

[@nocomments_s]

I like your video, it deserves much more views

[@brianmorin5547]

This is fantastic stuff as I’m pulling out my hair on this same step. You have the right idea for the next video, but I think the next one after that is making the chat interface publicly accessible

[@ShawhinTalebi]

Great suggestion Brian! There are several ways one can do this. The simplest and cheapest would be hosting it via HuggingFace Spaces: huggingface.co/spaces/launch However, for this specific use case the most practical option would be to embed it into my Squarespace website. I'll need to do some more digging to see the best way to do that.

[@brianmorin5547]

@@ShawhinTalebi The streamlit cloud has been my go-to so far. I am toying with creating a react front end template but would like to see what others are doing

[@pedro.henrique-k5z]

Hi! Great video! Thanks for sharing with us! Do you have any recommendation on what to study next to learn more about these topics or to go deeper in it?

[@ShawhinTalebi]

Happy to give recs. What's your end goal?

[@dhirajkumarsahu999]

Thank you so much, such videos are really very helpful

[@ShawhinTalebi]

Glad to hear :)

[@RatherBeCancelledThanHandled]

well done , Thanks !

[@fatimayousaf1644]

Hey Shaw! what a beautiful content. I followed all the steps from ML APP from scratch till deployment but at last moment, I don;t have AWS free tier account as they still ask to enter the debit card details for having free access, so can you please tell me another way round to cope up with this ?

[@ShawhinTalebi]

Glad you liked it! You could try deploying to railway (railway.app/). I just used them for a project and don't think I needed to input credit card info.

[@angieyoon9900]

Thank you!

[@paulohss2]

Great tutorial... Im still not able to connect to the API unfortunately (This site can’t be reached,refused to connect.) even though I followed the network config steps you explained...

[@ShawhinTalebi]

To confirm, you added inbound rules in the VPC dashboard to allow all incoming traffic from your IP? Does the IP listed in the inbound rules match yours?

[@thonnatigopi]

Bro create a video for handling post and get request and multiple endpoints using fast api dockerize and ECR and aws lambda functions

[@ShawhinTalebi]

Great suggestion. I'll add that to my list!

[@abbasrabbani7665]

that was an awesome video, I have a task for one click ml model deployment on aws, azure and GCP, like one click on aws and other click on azure. CAn u please guide me shortly the roadmap...!

[@ShawhinTalebi]

Thanks for your comment. Sorry I'm not sure what you mean by "one click ml model deployment". Could you share more details?

[@Smrigankiitk]

how can we integrate streamlit to make the UI to get input and send to model and display the output here ?

[@ShawhinTalebi]

Good question. This should be similar to the Gradio example shown 25:54. This blog post might be helpful: blog.streamlit.io/create-a-search-engine-with-streamlit-and-google-sheets/

[@ax5344]

where does K8S fit in in this pipeline?

[@ShawhinTalebi]

Good question. In my experience, Kubernetes is rarely used in DS/ML, so I wouldn't worry learning it if your just getting started.

[@rameshbabu2085]

I have a DL model which takes about 5 mins and 3gb GPU to process the query and to return result. I need to handle 5 queries per minute and I have a GPU with 8gb in GCP. How can I deploy such a model without memory leakage and I should be able to use the GPU at its full potential?

[@ShawhinTalebi]

How big is that model? Do you have GPU parallelization enabled? If it takes 5 min and 3GB to do one query with parallelization, the model may be too big to meet those technical constraints.

[@vivekasthana12345]

Thank you Shaw for making so many amazing videos. quick question from this video.. where exactly are you making a connection between your dockerhub and AWS ECS? Is it where you mention the url of the image? what if someone has a similar image name (shawhint/yt-search-demo) or that's not possible? Sorry if its a dumb question 😐

[@ShawhinTalebi]

Good question! Yes, exactly. No one will have a similar image name because the first part will be your unique DockerHub username.