A course where elementary students learn the basics of NumPy, Pandas, and scikit-learn and train their first ML model themselves. Students understand K-NN classification and linear regression through graphs and even experience model evaluation, so they enter the M1 middle-school comprehensive program (compressed F1+F2) not from a cold start but in a warmed-up state.
Published: May 16, 2026 | Last updated: May 16, 2026 ยท Based on the KOAI 2026 guidelines
Track
Elementary Pipeline
M1 Bridge
Target Grade
Grades 5-6
Late grade 4 for fast learners
Recommended Hours
~12 hours
1:1 basis ยท varies 8-16 hours
KOAI Mapping
M1 Warm-up
Prevents a cold start
By the time they finish E3, students will have learned the basics of NumPy, Pandas, and scikit-learn and trained their first ML model hands-on. Instead of difficult formulas, they use graphs and familiar data (weather, sports, games) to intuitively understand the principles of K-NN classification and linear regression, with the goal of reaching a level where they can explain "what accuracy means" on their own.
The core purpose of E3 is to enter the M1 middle-school comprehensive program (compressed F1+F2) in a warmed-up state rather than from a cold start. By the time they enter 7th grade, they have already worked hands-on with NumPy, Pandas, visualization, and model evaluation, so they start from a different line than students beginning M1 entirely from scratch.
E2. Elementary Python + Data completion is required. Students also need to be comfortable with the four arithmetic operations and the concepts of averages and ratios in order to keep up comfortably with the data and model evaluation. E3 doesn't force difficult formulas; it explains the principles through graphs and familiar data.
Below is the standard plan on a 1:1 basis. Depending on the student's prior knowledge and pace, some weeks may be accelerated, compressed, or covered in greater depth. Key tools: NumPy, Pandas, Matplotlib, intro to scikit-learn, kid-friendly Kaggle Learn notebooks, Teachable Machine code integration.
| Week | Topic | Key Deliverable |
|---|---|---|
| 1-3 | Working with numbers in NumPy (visual understanding of vectors and matrices) | NumPy mini notebook |
| 4-6 | Working with tables in Pandas (weather, sports, game data) | 1 EDA notebook |
| 7-9 | Matplotlib/Seaborn visualization | Data storytelling notebook |
| 10-12 | First ML model: classification with K-NN | KNN classification notebook |
| 13-15 | Linear regression (understood through graphs) | Regression mini project |
| 16-17 | Model evaluation: what accuracy means | Evaluation metrics notebook |
| 18-19 | Take on a kid-friendly Kaggle competition | 1 Kaggle submission |
| 20 | End-of-term presentation + M1 readiness prep | Capstone presentation |
โป Weeks are content units; actual time required varies by student. Recommended ~12 hours, ranging 8-16 hours.
Students create a NumPy mini notebook, an EDA notebook, and a visualization (data storytelling) notebook step by step. Working hands-on with familiar data, they internalize how to use the tools.
1 GitHub repo first-ml-projects to organize their results, leave behind 2-3 Kaggle notebooks, and explain their first ML journey themselves in an end-of-term presentation.
By finishing E3, students arrive at M1 with 2-3 years of history already accumulated on GitHub and a head start on Kaggle activity. This is a very rare strength within the applicant pool of their age group, and it directly affects the depth and specificity of the KOAI Round 1 application's "Portfolio 40%" score.
The Difference at M1 Entry
E3 graduates have already experienced about 5-6 weeks' worth of the first 14 weeks of M1 (compressed F1) โ NumPy, Pandas, visualization, model evaluation, and more. โ After a diagnostic, the M1 instructor can place them in an advanced section or run them at an accelerated pace.
GitHub
first-ml-projects repo
Kaggle
2-3 notebooks + first submission
Notion
Learning log
This track record accumulates as dated evidence in the KOAI Round 1 application's Portfolio 40% item. The earlier you start, the more depth you'll have by exam time.
E3 is the final step of the elementary pipeline and the bridge leading into the M1 middle-school comprehensive program. For the full track structure, see the KOAI Prep Curriculum Hub.
Current Course
E3. First ML for Elementary
NumPy, Pandas, first ML model
Recommended for grades 5-6 (late grade 4 for fast learners). It's designed for students who have completed E2 and are comfortable with the four arithmetic operations and the concepts of averages and ratios.
E3 is the bridge into the M1 middle-school comprehensive program. E3 graduates have already covered about 5-6 weeks' worth of the first half of M1 (compressed F1), so after a diagnostic they can be placed in an advanced section or run at an accelerated pace.
Students take on kid-friendly competitions for learning purposes with parental consent, and getting started on this activity early becomes a rare strength in the KOAI Round 1 application.
The M1 middle-school comprehensive program upon entering 7th grade. For exact KOAI exam dates, check the KOAI Competition Guide (https://citcoding.com/competitions/koai.html).
In a diagnostic session, we design an individualized bridge path that leads smoothly from E3 to M1.