Logistic Regression Lab Session
Dear students,
This thread is intended to collect your feedback, questions, and comments about Lab Session 3: Logistic Regression for Classification.
In this session, the material covers the Iris dataset, the logistic function for binary classification, decision boundaries, one-parameter binary classifiers, multiparametric classifiers, and multiparametric multiclass classifiers. The session also includes video materials and a repository with the scripts used during the lab.
Please use this post to share:
- Which part of the session was clear and useful.
- Which topic was difficult to understand.
- Whether the videos and repository were sufficient to follow the lab.
- Any errors, unclear explanations, or missing steps in the scripts.
- Suggestions to improve the next session.
Guiding questions
- Did you understand the difference between binary and multiclass logistic regression?
- Was the explanation of the decision boundary clear?
- Were you able to run the examples with the Iris dataset?
- Did the repository help you reproduce the results?
- What topic should be explained again with more detail?
Important references for this session
- Video: Part 1 — Logistic Regression
- Video: Part 2 — Multiparametric and multiclass session
- Repository: Logistic regressor](https://gitea.itmorelia.com/aiam/Logistic-regressor-video-p1.git)
These resources were shared as part of the lab session announcement.
Please write your feedback in a respectful and specific way. If you had difficulties, indicate exactly which part of the process caused the problem, for example:
- dataset loading,
- feature selection,
- gradient descent interpretation,
- sigmoid function,
- model training with scikit-learn,
- multiclass implementation,
- plotting or interpretation of results.
Your comments will be used to improve the next sessions.