๐Ÿฆ‰DVC Profile picture
Jan 16, 2023 โ€ข 7 tweets โ€ข 4 min read โ€ข Read on X
๐Ÿ‘จ๐Ÿปโ€๐Ÿ’ป Setup CI/CD in your machine learning projects using these simple yet powerful โ€œโ™พ๏ธ CML commandsโ€:

๐Ÿ”„ ci
๐Ÿƒโ€โ™‚๏ธ runner
โคด๏ธ pr
๐Ÿ’ฌ comment
๐Ÿง‘๐Ÿผโ€๐Ÿซ tensorboard

@Iterativeai @DVCorg
#cml #tensorboard #opensource

๐Ÿงต[1/7]
๐Ÿ”„ ci

The โ€œcml ciโ€ command prepares the whole repository for CML operations after creating the cml.yaml file

๐Ÿงต[2/7]
๐Ÿƒโ€โ™‚๏ธ runner

When a workflow requires computational resources, CML can automatically allocate cloud instances using โ€œcml runnerโ€. You can spin up instances on AWS, Azure, GCP, Kubernetes or any other provider. We can start a runner using โ€œcml runner launchโ€ command.

๐Ÿงต[3/7]
โคด๏ธ pr

We can also commit specified files to a new branch and create a pull request using โ€œcml pr createโ€ command.

๐Ÿงต[4/7]
๐Ÿ’ฌ comment

We can effortlessly post a Markdown report as a comment on a commit or pull/merge request with โ€œcml comment createโ€ command.

๐Ÿงต[5/7]
๐Ÿง‘๐Ÿผโ€๐Ÿซ tensorboard

The โ€œcml tensorboard connectโ€ returns a link to a tensorboard.dev page.

๐Ÿงต[6/7]
๐Ÿ‘จ๐Ÿปโ€๐Ÿ’ป Thanks for reading till here!

You can head straight to the docs to start integrating all these features in your projects ๐Ÿ‘‡
cml.dev/doc/ref

๐Ÿงต[7/7]

โ€ข โ€ข โ€ข

Missing some Tweet in this thread? You can try to force a refresh
ใ€€

Keep Current with ๐Ÿฆ‰DVC

๐Ÿฆ‰DVC Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @DVCorg

May 18, 2023
๐—–๐—ผ๐—บ๐—บ๐˜‚๐—ป๐—ถ๐˜๐˜† ๐—ฆ๐—ฝ๐—ผ๐˜๐—น๐—ถ๐—ด๐—ต๐˜!!

We were privileged to have Matt Squire from @FuzzyLabsAI, a renowned expert in the MLOps space, as our guest.

During the session, Matt shared his insights on the benefits of open-source MLOps tools and how they can help businesses.

๐Ÿงต๐Ÿงต Image
In his ๐—ฃ๐—ฟ๐—ฒ๐˜€๐—ฒ๐—ป๐˜๐—ฎ๐˜๐—ถ๐—ผ๐—ป, Matt provided a comprehensive breakdown of the MLOps tool space, categorizing it into SaaS platforms, fully open source, and partly open source tools.
He went on to describe the defining traits of open source and why he believes it to be the best choice for MLOps, including its flexibility, cost-effectiveness, and agility.

Matt's views on open-source MLOps tools generated a lot of engagement from our community members.
Read 4 tweets
May 17, 2023
Several data practitioners have asked for a tutorial video from us on DVC, and we are glad to make this available.

In this video, you'll discover how to use the DVCLive and @huggingface datasets to create a Tweet Sentiment Analyzer.

๐Ÿงต๐Ÿงต๐Ÿงต Image
Do you want to learn more about using real-time metrics in machine learning experiments? Look nowhere else!

In this ๐˜๐˜‚๐˜๐—ผ๐—ฟ๐—ถ๐—ฎ๐—น, we'll show you how to use DVCLive and the Hugging Face Dataset to build a potent Twitter sentiment analyzer.
๐ƒ๐•๐‚๐‹๐ข๐ฏ๐ž, a library from ๐ƒ๐•๐‚, gives you the ability to easily monitor your ML experiments. You will be able to easily grasp the performance of your model at every stage thanks to its real-time metrics.
Read 9 tweets
Apr 25, 2023
We are happy to announce something exciting to the ML/AI Community๐Ÿšจ

DVC has just released its new integration with Optuna, enabling you to streamline and optimize your hyperparameter search process while keeping track of every step with version control.
Our users have been requesting this integration for a while, and we're thrilled to deliver!

With the DVC's extension for VS Code, you can easily monitor and analyze your results, saving you time and effort in your machine learning workflow.
Try out our new integration today and see the benefits for yourself! ๐Ÿš€

This is the link to the video -

In the video, we explain how to use Optuna with Keras and view each iteration as an experiment in DVC's experiment table and plots. ๐Ÿ”๐Ÿ“ˆ
Read 4 tweets
Apr 20, 2023
Here's a comprehensive thread on how to build a seamless batch-scoring experience.

You will learn how to:

๐Ÿค– Design ML pipelines with @DVCorg

๐Ÿˆ Training with @gitlab and CML

๐Ÿงช ML experiments with @Iterativeai Studio

๐Ÿšข Deployments with @ApacheAirflow

๐Ÿงต [1/6]
๐Ÿค– Design #ML pipelines using #DVC! Streamline your process with the following steps:

1๏ธโƒฃ Data prep

2๏ธโƒฃ Feature engineering

3๏ธโƒฃ Model training

4๏ธโƒฃ Model evaluation

DVC optimizes run time & tracks changes for a more efficient workflow!

Get started with 'dvc exp run' ๐Ÿš€

๐Ÿงต [2/6]
Read 14 tweets
Feb 28, 2023
What is your favourite IDE/Code Editor for Machine Learning?

1. Jupyter Lab/Notebook
2. R Studio
3. Pycharm
4. VS Code
5. Spyder
6. Others(name it)
Follow @DVCorg

A lot of people keep mentioning @code

Check out this @code Extension for Machine learning experiment management -

marketplace.visualstudio.com/items?itemNameโ€ฆ
These are the Top @code extensions for Machine learning in 2023.

1. GitHub Copilot
2. DVC
3. GitLens
4. Jupyter
5. Jupyter Notebook Renderers
6. Python
7. Pylance
8. Python Indent
9. Indent-rainbow
10. Todo MD

You can also share other ones to educate others.
Read 5 tweets
Feb 13, 2023
Woah! Been here? Is deep learning model training going horribly wrong? ๐Ÿ™‹๐Ÿฝโ€โ™‚๏ธ

Iterative Studio makes this easy to see so you don't waste time and resources!

๐Ÿงตย 1/7
With Iterative Studio and DVCLive, you can monitor the progress of your long-running experiments against others that you or your team have performed. All are easily accessed at work, at home, or by the rest of your team on the project.

๐Ÿงต2/7
You provide a couple environment variables for your model training job:

You can enter your STUDIO_TOKEN and dvc exp run if running locally

๐Ÿงต3/7
Read 7 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(