Skip to content

Track 01

Python, NumPy, Pandas, Visualization

This first track turns data handling into a real workflow: inspect the shape, inspect the table, inspect one slice, inspect one plot, then build one feature matrix without losing row alignment.

Start With Tooling Topics Run Fast Examples Use The Study Plan

Primary Goal

Inspect Before Modeling

The student should leave this track able to see data problems early instead of discovering them later through a broken model.

Best For

First-Time Learners

This is the default first track when arrays, tables, grouped summaries, and plotting are not yet mechanical.

Exit Rule

Leave With One Clean Baseline Table

You are done when the data path feels stable enough that the first model is no longer the scary part.

Use This Track When

  • you are learning AI for the first time
  • table inspection is still slow
  • grouped summaries and counts do not yet feel automatic
  • feature construction still risks row misalignment

What This Track Is Training

This track trains one durable habit:

  • inspect before you transform

That means the learner should be able to move through this order without rushing:

  1. shapes
  2. schema
  3. missingness
  4. slice counts
  5. one useful plot
  6. one stable feature matrix

First Session

Use this order:

  1. Array Shapes and Axis Operations
  2. Table Inspection
  3. run academy/.venv/bin/python academy/examples/numpy-shape-sanity/shape_sanity.py
  4. run academy/.venv/bin/python academy/examples/course-support-inspection/quick_inspection.py
  5. write one short note on the first suspicious slice or feature risk

Full Track Loop

For the complete workflow:

  1. read the tooling topics in order
  2. run one short example after each core topic
  3. run academy/.venv/bin/python academy/labs/python-numpy-pandas-visualization/src/course_support_eda.py
  4. run academy/.venv/bin/python academy/labs/python-numpy-pandas-visualization/src/feature_matrix_walkthrough.py
  5. finish Exercise Sets 1 and 2 in academy/exercises/python-numpy-pandas-visualization/ before moving to the tiny logistic baseline in Set 3
  6. keep one note with the baseline table, slice summary, and first feature-matrix checks

What To Inspect

By the end of the track, the learner should have inspected:

  • row and column counts
  • missing-value rates
  • target balance or label counts
  • at least one grouped summary with counts
  • one plot that changes the next action
  • one feature matrix whose row alignment is still explicit

Common Failure Modes

  • moving to modeling before the schema is understood
  • trusting a slice rate without its count
  • using plots decoratively instead of diagnostically
  • building derived features without checking row alignment
  • converting to a matrix too early and losing the meaning of columns

Exit Standard

Before leaving this track, the learner should be able to:

  • explain what the table contains
  • name the first risky slice
  • build one clean feature matrix
  • say what should be inspected before the first baseline model

That is enough to move safely into scikit-learn Validation and Tuning.