Kishor.Academy
Log in
← All courses

Artificial Intelligence · beginner

Python & Data for Machine Learning

Learn practical Python, CSV inspection, data validation, reproducible splits, and leakage prevention through a complete workshop-data lab.

With Admin

beginner9 lessons · 3 sections8–12 hours of reading and coding, plus 3–4 hours for the project; estimates exclude video.

THE TAKEAWAYS

What you’ll learn

  • Run a Python script and distinguish numeric values from text.
  • Use lists, dictionaries, loops, and functions to process small records.
  • Read CSV rows, inspect errors, and record reproducible commands.
  • Explain tabular data and inspect columns, types, missingness, and summary statistics.
  • Validate bad counts and audit exact duplicates and conflicting identifiers.
  • Choose a useful visualization and distinguish a visible pattern from a causal claim.
  • Define features and labels at a prediction cutoff and identify leakage.
  • Create reproducible train/test splits and explain when temporal or grouped splitting is needed.
  • Deliver a data dictionary, cleaning audit, split evidence, and runnable data report.

Is this for you?

Beginning programmers and aspiring ML learners who need a reliable data foundation before training models.

Before you start

  • A computer on which you can run Python 3.10 or newer and a text editor.
  • Basic arithmetic and willingness to read error messages; prior programming is not required.
  • Download the lab bundle and preserve its folder structure. No API key or paid notebook required.

PUT IT TO WORK

What you’ll build

Reproducible workshop data report. Client scenario: an organizer needs a trustworthy workshop table before prediction. Deliverable: a data report, audit, data dictionary, and reproducible commands.

Move from values, lists, dictionaries, loops, and functions to a reproducible data workflow. Use a small fictional workshop dataset whose errors are visible: missing values, invalid counts, duplicates, and conflicting identifiers. Record an audit rather than silently changing data. Build a data dictionary, split by the right unit, and explain why a field available after the prediction cutoff is not an eligible input. All required code uses the Python standard library.

Course curriculum

SECTION 1Python building blocks and tabular data3 lessons
  1. Notebook setup and your first Python programPython practice
  2. Lists, dictionaries, loops, and functionsPython practice
  3. Files, errors, and reproducible codePython practice
SECTION 2Validate, clean, and explain the data3 lessons
  1. DataFrames and exploratory analysisPython practice
  2. Cleaning data without hiding decisionsPython practice
  3. Visualizing patterns and asking better questionsPython practice
SECTION 3Honest splits, reproducibility, and a data report3 lessons
  1. Features, labels, and data leakagePython practice
  2. Train, validation, and test splitsPython practice
  3. Project: prepare a real datasetPython practice