Best MLOps Practices for Building End-to-End Machine Learning Computer Vision Projects with Alex Kim

In this workshop with DataTalks Club, we’ll build an end-to-end Computer Vision system using MLOps tools DVC, CML, the DVC extension for VS Code, and Iterative Studio along with Fast AI, nvtop and Docker.

We’ll explore an industrial use case of training an image segmentation model for the purposes of defect detection on a manufacturing conveyor belt.

The dataset and the use case are described in this [paper](https://www.researchgate.net/profile/....
Repo: https://github.com/iterative/magnetic...
*******
You’ll learn:

How to quickly configure a remote development environment with [TPI](https://github.com/iterative/terrafor... write code locally while executing on a remote machine with a GPU
How to version large datasets and models with [DVC](https://github.com/iterative/dvc)
When it’s the right time to move from Jupyter notebooks to ML pipelines and how to do that with [DVC](https://github.com/iterative/dvc)
Why it’s beneficial to integrate CI/CD workflows into your model development process and how to do that with [CML](https://github.com/iterative/cml)
How to manage experiments and collaborate on ML projects using [Iterative Studio](https://studio.iterative.ai/)
********
Target Audience:

Technical folks (e.g. Software Engineers, ML Engineers, Data Scientists) who are familiar with general Machine Learning concepts, Python programming.

Knowledge of CI/CD processes and Cloud infrastructure will be helpful.
********
Prerequisites:

AWS Account: [https://mlbookcamp.com/article/aws](https://mlbookcamp.com/article/aws)
Familiarity with AWS S3 and AWS EC2
Familiarity with GitHub Actions will be helpful: [https://github.com/features/actions](https://github.com/features/actions)
********
About the speaker:

Most of Alex’s work experience involved solving data science problems in various domains: physics, aerospace, telemetry/log analytics, image, and video processing.

In the last couple of years, he became increasingly interested in the engineering side of ML projects: processes and tools needed to go from an idea to a production solution. Currently, he works as an MLOps Solutions Engineer at [Iterative.ai](http://iterative.ai/), helping customers extract the most value from the Iterative ecosystem of tools.

Short video trailers of what will be covered in the talk

[1 - Launch VSCode with TPI.mp4](https://drive.google.com/file/d/1yl8F...)

Shows how to achieve local development experience while running code on a Cloud machine with a powerful GPU.

[2 - Create DVC pipeline.mp4](https://drive.google.com/file/d/1pF3I...)

Introduces DVC pipelines

[3 - DVC and VSCode Extension.mp4](https://drive.google.com/file/d/1jdXo...)

Shows how to manage experiments with VSCode Extension for DVC

[4 - CICD and CML.mp4](https://drive.google.com/file/d/1DPpG...)

Shows how to configure CI/CD jobs powered by CML: deploying cloud runner and automatic reporting to GitHub.

[5 - Exp Management in Studio.mp4](https://drive.google.com/file/d/1nKJo...)

Experiment management in Studio, plots, running remote experiments via Studio UI

Try out the DVC Extension for VS Code here: https://marketplace.visualstudio.com/...

To learn more about Iterative's open-source and SaaS tools please visit:
🧑🏽‍💻 Our online course: https://learn.iterative.ai
✍🏼 Our docs: https://dvc.org/doc (Data Version Control, Pipelines, Experiments)
https://cml.dev/doc (CI/CD for Machine Learning)
https://mlem.ai/doc (Package and Serve your models)
https://studio.iterative.ai (Team Collaboration, Experiments, Model Registry)

Join our Discord server:   / discord  

#dvc #machinelearning #datascience