visenger/awesome-mlops

A curated list of references for MLOps

awesome-mlops's Issues

Add Deep-Learning-in-Production

Hi,
Thanks for your great repo. You can also add the below repo to your list.
https://github.com/ahkarami/Deep-Learning-in-Production

Best

Click to expand!

Click to expand!

These make it difficult to read the content.

Devops

Suggestion : Automate your cycle of Intelligence

Katonic MLOps Platform is a collaborative platform with a Unified UI to manage all data science activities in one place and introduce MLOps practice into the production systems of customers and developers. It is a collection of cloud-native tools for all of these stages of MLOps:

-Data exploration
-Feature preparation
-Model training/tuning
-Model serving, testing and versioning

Katonic is for both data scientists and data engineers looking to build production-grade machine learning implementations and can be run either locally in your development environment or on a production cluster. Katonic provides a unified system—leveraging Kubernetes for containerization and scalability for the portability and repeatability of its pipelines.

It will be great if you can list it on your account

Website -
Katonic One Pager.pdf

https://katonic.ai/

Free MLEngineering and Model Deployment courses and repo

Check and Add this free courses to your awesome mlops list if you find it relevant. These courses are created by me

MLEngineering - https://www.youtube.com/playlist?list=PL3N9eeOlCrP6Y73-dOA5Meso7Dv7qYiUU

Model Deployment - https://www.youtube.com/playlist?list=PL3N9eeOlCrP5PlN1jwOB3jVZE6nYTVswk

Model Deployment on Google Cloud Platform - https://www.youtube.com/playlist?list=PL3N9eeOlCrP4VXtFJTjmGsqI-Emk2keVL

You can find code for these videos in my git repo

New paper: "A software engineering perspective on engineering machine learning systems: State of the art and challenges"

I kindly propose to have a look at the paper: A software engineering perspective on engineering machine learning systems: State of the art and challenges. Link @ScienceDirect, Link @Arxiv

[Suggestion] Computer Vision Newsletter

I curate a bi-weekly community newsletter for computer vision practitioners. No marketing or product in it, just CV news, articles, learning resources, MLOps & DataOps, events, etc. It is free and focused on bringing value to the AI community.
I would love to add it to the awesome-mlops repository.

Here is the subscription link
Here is one of the last issues to check out.

Fixed-link : Model Deployment and Serving

Fixed the links in point 12 in the MLOps: Model Deployment and Serving section. Kindly check and merge. Thanks. :)

Reformatting the MLOps section

Hi,

In a previous PR, I suggested simplifying the section "MLOps," which IMHO looks a little confusing, given that it follows different citation styles.
In response to @visenger question (Would you take the lead and "chicago-fy" the paper's references?), yes, I can do that.
However, on the one hand, I wonder whether adding authors and venue information is really needed, considering that a large part of the text is made of that information, and the remaining is the title, which is the most useful information.
On the other hand, most of the list items in the other sections have an elementary format given by their titles only.

That being said, I would like to propose two alternatives to reformat the MLOps section:

Alternative 1

Reformat each item using the title only, and link it to the raw document or its digital library page (if available). For example

MLOps Papers

1. Challenges in deploying machine learning: a survey of case studies.
2. Challenges in the deployment and operation of machine learning in practice.
3. On challenges in machine learning model management.

Alternative 2

Reformat each item using the title plus a brief description from its abstract, and link it through a "Go to paper" to the raw document or its digital library page (if available). As here. For example:

MLOps Papers

1. (2021) Challenges in deploying machine learning: a survey of case studies. This survey reviews published reports of deploying machine learning solutions in a variety of use cases, industries and applications and extracts practical considerations corresponding to stages of the machine learning deployment workflow. Go to paper.
2. (2019) Challenges in the deployment and operation of machine learning in practice. In this work, the authors target to systematically elicit the challenges in deployment and operation to enable broader practical dissemination of machine learning applications. Go to paper.
3. (2018) On challenges in machine learning model management. This paper discusses a selection of ML use cases, develops an overview over conceptual, engineering, and data-processing related challenges arising in the management of the
   corresponding ML models, and points out future research directions. Go to paper.

Alternative 1 is consistent with the rest of the README; while alternative 2 provides more details about each paper without being too difficult too read. In summary, I believe it helps better identify the papers of interest.

While I agree to give credit to the authors of the papers, they are listed in the papers themselves and on the digital library page. Therefore, I found it either redundant and cumbersome.

Note: I do not want to dictate new rules on how to format the document. I simply find that this way would be more effective to identify the resources of interest for a given search, at least for me :)

Of course, I will be in charge of those modifications.

What do you think?

MLOps talk

Please add this talk,

Emmanuel Raj - MLOps: Automated Machine Learning | PyData: https://www.youtube.com/watch?v=m32k9jcY4pY

Formatting issue

Hi, in the machine learning section, there is a formatting issue:

Line 219-220, issue: a missed space

I would be happy If I could send a PR.

Thanks for great work.

Engineering MLOps book

Please add the book,

"Engineering MLOps: Rapidly build, test, and manage production-ready machine learning life cycles at scale" by Emmanuel Raj, 2021 - https://www.packtpub.com/product/engineering-mlops/9781800562882

Soon to update another MLOps link

How are the diagrams created?

Hi @visenger
Thank you for the amazing resources. Learnt a lot from them. May I know what tool you are using to create the diagrams

New Paper

https://www.dropbox.com/s/wa6chu5qs97a49a/ad-pulse-ai_v3.pdf?dl=0

This explains how to make a regressive network from a few simple examples. By taking what one might call a polyhedron of angles of the object, the source picture behaves as a thesaurus of the standing class of the object. Thus, it should be easily recalled by a TensorFlow AI.

This is my self-written AI. I just don't have access to the pictures like in need in the paper.
https://github.com/wise-penny/LIDSx