MLP Live session Week 7

MLP Live session Week 7

by Machine Learning Practice

Applied Sciences & Engineering AI — Models & Research UYQArtificial intelligence (AI)UYQNNeural networks and fuzzy systems
🎙 Machine Learning Practice 👥 4K 📅 July 28, 2026 ⏱ 90 min 👁 228 📄 tutorial 🧭 2026-08-18
Available in: English (current) Français

Keywords

OPPKagglepre-processingmodel buildingpandas

Summary

This live session is a Q&A and guidance session for the Machine Learning Practice (MLP) course, led by an instructor. The session primarily addresses student questions about the upcoming OPP (Open Book Practical) exam and the second Kaggle assignment. The instructor clarifies that the first Kaggle assignment will be graded partially, with marks given for the submitted component (Kaggle part) even if the peer review is missed, but this leniency will not apply to future assignments. He explains the OPP exam structure: it consists of two sections - pre-processing and model building - with approximately 18-20 questions total, to be completed in 1.5 hours. Pre-processing will involve cleaning a dataset (handling missing values, scaling, encoding, etc.) and answering questions using pandas, while model building will require building models, pipelines, and hyperparameter tuning on a clean dataset. The instructor advises students to practice with week 1-3 assignments for pre-processing and week 5 assignments for model building, and to use datasets like California housing or wine. He emphasizes that no external documentation is allowed, but the help() and dir() functions in Python are permitted, and AI features in Colab will be disabled. He also mentions that the second Kaggle assignment may be released after the OPP, depending on support team approval. The session also addresses administrative issues like missing OPP slots and email communication.

222 words

Critical Evaluation

Value of the Information & Strength of the Argument

The session provides valuable, practical information for students preparing for the OPP exam and managing course assignments. The instructor gives clear, actionable advice on exam structure, preparation strategies, and common pitfalls (e.g., using random_state, avoiding AI features). The argumentation is straightforward and based on the instructor’s experience and course policies, not on empirical evidence or external sources. The advice is consistent with standard machine learning practices, making it reliable for the intended audience.

Scientific Rigor, Source Quality, Title Accuracy

The session does not cite external sources or references; it relies on the instructor’s knowledge and course materials. The information is internally consistent and aligns with typical ML course content. The title accurately reflects the content, as it is a live session for the MLP course. The lack of formal citations is expected for a tutorial-style Q&A session, but it limits the scientific rigor. The session does not present original research or data, but rather practical guidance.

165 words

Title / Content Match

The title accurately reflects the content: a live session for the MLP course, focusing on week 7 topics including OPP exam details and Kaggle assignment clarifications.

Quality & Reliability

6/10

The session is an instructor-led Q&A and guidance session for a machine learning course, providing practical advice on exam preparation and assignment logistics. Information is based on the instructor's knowledge and course policies, not on external research or data. The advice is practical and consistent with standard ML practices, but lacks formal citations or references.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Contribution & Novelties

The session provides practical, course-specific guidance that is not typically found in textbooks or online tutorials. It offers a clear breakdown of the OPP exam structure and preparation strategies, which is valuable for students in the course. The advice on using help() and dir() functions and avoiding AI features is specific to the exam environment.

Pour aller plus loin :

  • Scikit-learn documentation — Official documentation for scikit-learn, useful for understanding model building and preprocessing tools.
  • Pandas documentation — Official documentation for pandas, essential for data manipulation and preprocessing.
  • Kaggle — Platform for data science competitions, relevant for the Kaggle assignments mentioned in the session.

104 words

Radar Profile

The radar profile shows moderate scores across all dimensions, indicating a balanced but not exceptional session. The quantity and quality of information are adequate for a tutorial, but the lack of external sources and original research limits the scientific depth. The technical level is appropriate for the target audience, and the reliability is consistent with instructor-led guidance.

Reliability 6/10