AI-ML with Python
16-Week Complete Roadmap

Week 01Foundation

Google Colab + Python start

Introduction to Colab, cells, markdown, running code, variables, input-output, data types, operators, Python editor, and VS Code awareness.

• Learn: notebook workflow, syntax, errors, print, comments, editor basics, and VS Code setup awareness
• Practice: simple calculator and logic tasks inside Colab
• Output: first personal Colab notebook and first runs in the Learn With Champak editor

Week 02Foundation

Conditions, loops, functions, and strings

Flow control, reusable logic, string handling, and problem-solving patterns.

• Learn: if-else, for, while, functions, return, string methods
• Practice: text cleaning and menu-based mini programs
• Output: utility notebook with reusable functions

Week 03Foundation

Lists, dictionaries, NumPy, and arrays

Data structures plus the move from plain Python collections to numerical arrays.

• Learn: indexing, slicing, dictionaries, NumPy arrays, shape, axis
• Practice: compute mean, median, min, max, range with arrays
• Output: basic numerical thinking notebook

Week 04Foundation

pandas and reading CSV files in Colab

Series, DataFrame, CSV upload, selection, filtering, and first dataset inspection.

• Learn: read_csv, head, tail, info, describe, filtering
• Dataset: small classroom CSV + Titanic preview, explored inside Colab and editor practice
• Output: first data inspection report

Dataset Links Titanic CSV

Week 05Data core

Data cleaning and preparation

Missing values, duplicates, datatype fixes, string cleanup, and making data analysis-ready.

• Learn: null handling, fillna, dropna, duplicate control, type conversion
• Dataset: Titanic or messy student dataset
• Output: cleaned dataset notebook with before/after comparison

Dataset Links Titanic CSV

Week 06Data core

Visualization and insight building

Bar charts, histograms, scatter plots, box plots, and how to tell a simple story from data.

• Learn: matplotlib basics and chart interpretation
• Dataset: Titanic, housing, or sales-style CSV
• Output: chart-driven short analysis notebook

Dataset Links Titanic CSV Sales / Tips CSV MPG Regression CSV California Housing

Week 07Data core

Statistics for AI-ML

Mean, median, mode, variance, standard deviation, percentiles, distribution basics, and interpretation.

• Learn: descriptive statistics from ground level
• Practice: compute stats with NumPy and pandas
• Output: statistics worksheet notebook

Dataset Links Tips / Stats CSV Titanic CSV

Week 08Data core

Correlation, probability, and dataset selection

Understand relationships between variables and learn how different datasets support different problems.

• Learn: correlation matrix, simple probability intuition, target vs features
• Datasets: Iris, housing, wine overview
• Output: dataset comparison notes in Colab

Dataset Links Iris CSV sklearn Iris sklearn Wine California Housing

Week 09Similarity + NLP

Text data basics and preprocessing

Move into NLP with tokenization, lowercasing, punctuation removal, stopword awareness, and text cleanup.

• Learn: basic NLP pipeline from raw sentence to cleaned tokens
• Dataset: SMS Spam Collection small samples
• Output: text preprocessing notebook

Dataset Links SMS Spam Collection

Week 10Similarity + NLP

Bag of Words, TF-IDF, and cosine similarity

Represent text numerically and measure how similar two texts are.

• Learn: CountVectorizer, TF-IDF idea, vectors, dot product intuition, cosine similarity
• Practice: compare sentences, articles, or titles inside Colab
• Output: mini text similarity notebook

Dataset Links SMS Spam Collection 20 Newsgroups

Week 11Similarity + NLP

NLP mini applications

Use cosine similarity and vectorization in useful beginner projects like recommendation, matching, and search ranking.

• Learn: document matching, FAQ matching, basic recommendation logic
• Dataset: custom notes corpus or small movie/product text dataset
• Output: text matching mini project

Dataset Links 20 Newsgroups SMS Spam Collection

Week 12Similarity + NLP

Text classification introduction

A gentle bridge from text processing to machine learning using spam or sentiment-style tasks.

• Learn: train-test split for text, vectorize then classify
• Datasets: SMS Spam Collection, IMDB subset, or 20 Newsgroups sample
• Output: first NLP classification notebook

Dataset Links SMS Spam Collection 20 Newsgroups

Week 13ML core

Regression from ground zero

Targets, features, prediction, train-test split, fit, predict, error, and interpretation.

• Learn: linear regression and evaluation basics
• Datasets: housing or marks-prediction style dataset
• Output: regression notebook with charts and findings

Dataset Links California Housing MPG Regression CSV

Week 14ML core

Classification with important benchmark datasets

Modeling category prediction problems with classic datasets that are beginner-friendly and respected.

• Learn: logistic regression, k-NN, accuracy, confusion matrix
• Datasets: Iris, Wine, Breast Cancer Wisconsin, Titanic
• Output: classification comparison notebook

Dataset Links Iris CSV Wine Dataset Breast Cancer Wisconsin Titanic CSV

Week 15ML core

Clustering and unsupervised learning awareness

Introduce grouping without labels and help students understand the difference between supervised and unsupervised learning.

• Learn: k-means intuition, scaling awareness, cluster interpretation
• Datasets: Iris features or customer-style dataset
• Output: clustering notebook with visual interpretation

Dataset Links Iris CSV sklearn Iris

Week 16Capstone

Mini capstone in Google Colab

Choose a path and build something presentable: analysis, prediction, or text similarity / NLP mini app notebook.

• Project options: analysis dashboard notebook, spam classifier, recommendation / matching notebook, or regression/classification project
• Deliverables: Colab link, explanation, screenshots, conclusions
• Output: portfolio-ready final notebook project

Dataset Links Titanic CSV Iris CSV SMS Spam Collection California Housing 20 Newsgroups