Course Contents

The course is broadly divided in 7 categories, each of the topic is present as a section in the course.

Part 1. Introduction to Machine Learning

1. Python Recap
2. Intermediate Python
3. Machine Learning Introduction
4. Data Generation & Visualisation
5. Linear Algebra in Python

Part 2. Supervised Learning Algorithms

• Linear Regression
• Locally Weighted Regression
• Multivariate Regression
• Logistic Regression
• K-Nearest Neighbours
• Naive Bayes
• Support Vector Machines
• Decision Trees & Random Forests

Part 3. Unsupervised Learning

• K-Means
• Principal Component Analysis
• Autoencoders(Deep Learning)
• Generative Adversial Networks(Deep Learning)

Part 4. Deep Learning

• Deep Learning Fundamentals
• Keras Framework, Tensorflow Basics
• Neural Networks Basics
• Building Text & Image Pipelines
• Multilayer Perceptrons
• Optimizers, Loss Functions

Part 5. Deep Learning in Computer Vision

• Convolution Neural Networks
• Image Classification Pipeline
• Alexnet, VGG, Resnet, Inception
• Transfer Learning & Fine Tuning

Part 6. Deep Learning Natural Language Processing

• Sequence Models
• Recurrent Neural Networks
• LSTM Based Models
• Transfer Learning
• Natural Lang Processing
• Word Embeddings
• Langauge Models

Part 7. Reinforcement Learning

• Basics of Reinforcement Learning
• Q Learning
• Building AI for Games

Problem statements and mini-projects done in the course are:

• Hardwork Pays Off (Regression Prediction)
• Air Quality Prediction (Multivariate Regression)
• Separating Chemicals (Logistic Regression)
• Face Recognition (OpenCV, K-Nearest Neighbours)
• Handwritten Digits Classifier
• Naive Bayes Mushroom Classification
• Movie Review Prediction (Naive Bayes, LSTM etc)
• Image Dominant Color Extraction (K-Means)
• Image Classification using SVM
• Titanic Survivor Prediction using Decision Trees
• Diabetic Patients Classification
• Non-Linear Data Separation using MLP
• Pokemon Classification using CNN, Transfer Learning
• Sentiment Analysis using MLP, LSTM
• Text/Lyrics Generation using Markov Chains
• Emoji Prediction using Transfer Learning & LSTM
• Odd One Out (Word2Vec)
• Bollywood Word Analgoies (Word Embeddings)
• Generating Cartoon Avatars using GAN's (Generative Adversial Networks)
• Reinforcement Learning based Cartpole Game Player

Libraries, Frameworks

• Most of the course codes are build from scratch and also following libraries are used.

1. Pandas (Data Handling)
2. Matplotlib (Data Visualisation)
3. Numpy (Maths)
4. Keras (Deep learning)
5. Tensorflow(Introduction)
6. Sci-kit Learn(ML Algorithms)
7. OpenAI Gym (Reinforcement Learning)