article.md

Article (optional)

The best way to learn about something - to teach it.

In this part of the course, we'll explore some topics that weren't covered in the course.

You'll need to learn about this topic and then write about what you learned in an article. (Of course if you already know something and just want to share your knowledge with us - it's also fine and more than welcome.)

Articles from 2022

• https://medium.com/@sotoblanco263542/track-your-machine-learning-experiments-with-w-b-d5f9431e6bc2
• https://blog.aaishamuhammad.co.za/posts/onnx_machine_learning/
• https://github.com/ElenaNKn/scaling_methods
• https://medium.com/@alexandervalverdeguillen/math-optimization-methods-for-machine-learning-8837eba9e3fe
• https://rzabolotin.hashnode.dev/deploying-ml-model-via-telegram-bot
• https://github.com/bsenst/mlbookcamp-2022/blob/main/social-media-analysis-ml-zoomcamp.ipynb
• https://github.com/mary435/Telegram.git

Topics

You can write about pretty much anything that wasn't covered in the course, but related to machine learning and/or engineering.

Here are some ideas:

• Areas of Machine Learning:
  • Time series modelling
  • Machine learning for textual data
  • Recommender systems
  • Clustering
  • Dimensionality reduction
  • GANs
• Tools similar to what we covered:
  • Using poetry instead of pipenv
  • LightGBM and catboost instead of xgboost
  • Using PyTorch instead of Tensorflow
  • PyCaret
  • FastAI
  • TorchServe instead of TF Serving
• Some other interesting tools:
  • Kedro
  • Pandera
• Evaluation metrics:
  • Precision/Recall curves
• Pretty much everything from "explore more" sections
• And more!

It can be as broad as you'd like - for example, about time series in general or about some specific methods there (exponential smoothing).

If you're not sure if your topic is suitable or not, feel free to ask about it in Slack.

Requirements

For the topic, the only requirement is that it wasn't covered in the course, or not covered with the same amount of details that you want to have.

When learning the topic, please take a note of all the resources that you used and include them at the end in the "sources" section.

Where can I publish it?

• Medium
• Github
• Github pages
• Wordpress
• Your own blog

The top voted articles can be re-publised on the DataTalks.Club website if you'd like.

Plagiarism

• Copying from others fully or in part is not allowed.
• If you want to quote something from another article, mark it as a quote explicitly and point to the source immediately after the quote.

Re-using your own work

Allowed

Re-using your own work is okay if it's something that you did as a part of this course.

Examples:

• You wanted to deploy something to Heroku, you learned about it and wrote a tutorial
• You wanted to learn about PyTorch and used it for your project
• You did a project about time series and want to share what you learned

Not allowed

It's not allowed to re-use articles that you wrote before the course started.

Examples:

• You were taking part in another course and wrote an article for it. Now you want to re-use the same article here

Note: you're more than welcome to share this article with the community (e.g. in the #shameless-content channel in DataTalks.Club), but this article can't be submitted as an article for ML Zoomcamp.

The publication date for the article must be after the start of the course - September 2022.

Submitting the results

• Submit your article to #course-ml-zoomcamp-articles - just share the link to your article there
• Submit the same link to this form so we could link it to your message from the channel

Evaluating

We'll use voting for scoring your articles.

• Check the articles in the #course-ml-zoomcamp-articles channel and put a 👍 reaction to articles that you liked
• The top voted articles will get 20 points

Deadline

The deadline for finishing the article is 30 January, 22:00 CET.

Past articles

See examples of articles from the 2021 cohort here