The course offers a practical, hands-on experience in Machine Learning using Python. It includes comprehensive guidance on the entire ML pipeline, with many real-world applications. The provided notebooks come with detailed instructions, and some labs are supplemented with video lectures for additional support.
Course Outline
Week - 01
Overview of Python programming language
list, tuple, string, dictionary
alias, map, function
A brief introduction to other popular libraries
Week - 02
Python for Data Science
Numpy array
Pandas
Data Visualization
Matplotlib, Seaborn
Week - 03
Exploratory Data Analysis (EDA)
Importing different types of data
Identify issues in data - missing values, outliers, unrelated features
Missing data handling - imputation
Feature Engineering
Outlier removal
Open Ended Lab - 01 on EDA
Week - 04
Build your first ML model
Introduction to Scikit-learn
Data Preprocessing: imputation, normalization, removal
Different ways of validation: cross-validation, stratification
Model training, making and analyzing predictions
Example - 01: Regression problem (House pricing)
Example - 02: Classification problem
Week - 05
ML Pipeline
Validation Schemes
Different performance measures
Feature selection techniques - Wrapper methods
Hyperparameter tuning
Week - 06
Imbalanced Data handling
Data resampling techniques
Cost-sensitive Learning
Feature Selection
Filter methods
Real-world Application
Cervical Cancer Diagnosis
Week - 07
End-to-end ML pipeline with two real-world datasets
Open Ended Lab - 02 on ML model
Week - 08
Introduction to deep learning
Introduction to TensorFlow
Build your first DL/CNN model using TensorFlow
CIFAR dataset
Cat-vs-dog dataset
Week - 09
Working on real-world datasets
Hand-drawn Electric circuit Schematic Components prediction