Nov 12, 2024

Python
- Two Python Data Processing resources to consider these books
(Fast.ai)
- Completed the recommended intro lesson on NLP for lesson 4 https://www.kaggle.com/code/jhoward/getting-started-with-nlp-for-absolute-beginners
- How (and why) to create a good validation set
- Iterate like a grand champion
- Completed Lesson 4 - video
Google Colab
- Learned that all Google Colab notebooks for this course are linked here! but after using Colab, I found their ‘free’ tier to be quite limiting compared to what Kaggle provides. I’m sticking with Kaggle for now.
(Standford)
- Non-linear Features
  - Linear in what? Shows how the w*phi(x) is still linear, even when x is non-linear. This worked for both regression and classification. *Feature Templates
  - Example with an email address, and how you can define features with templates using either dense pre-defined arrays, or feature dictionaries that get defined dynamically (e.g., only non-zero features…which is ‘endsWith_com’ for one email but would be ‘endsWith_org’ for another)
Post two more blog articles.