Deployment

Hi everyone!

I’ve happily completed my machine learning specialization. Many thanks to Andrew for that!

Now I have built a three dockerized container app (model llama-2-13b-chat.Q4_K_M, backend, frontend) and now want to deploy it to AWS.

I’ve quickly understood that deployment is a big deal. Which service should I choose? What machine? How do you optimize pricing?

It seems that I’ll have to take another course. Which one should it be?

I will also be grateful for some immediate advice on how to deploy my project in a cost-efficient manner. Don’t want it to collect dust :slight_smile:

Cheers

1 Like

Hi @praxis

Welcome to the community.

Congratulations on complete the machine learning specialization.

Well, for a home/study project i would rather use dagshub or huggingface to deploy it.

It is free cost at certain point and provide a lot of features to maintain and develop a ML project

Hi @elirod

It’s a production-ready project, POC stage, so I want to publish it and have some user interactions.

There is a Deep Learning.AI course on machine learning production ops and deployment.

Thanks for the reference @TMosh . I’ve checked it out. Can you please clarify which platforms (AWS, Google, etc) are covered in this course? It doesn’t appear to be stated in the description.

Sorry, I do not know. I have not attended the course myself.

You might try reading through the forum threads for that course.

Own, i got it!

In that case, Oracle is a simple and low-cost choice from my point of view.