©2018 by Deeper Side of Learning. Proudly created with Wix.com

Machine Learning: Zero To Deep Dive

Class Syllabus

Part 1 - Introduction to Machine Learning:

 

 

 

 

 

Week 1: Introducing Data Science

  • Introduction

    • History of AI

    • What is Data Science

    • Machine Learning Applications

      • Early

      • Current

      • Future

    • Types of Machine Learning

  • Linear Algebra

    • Structures

      • Scalar, Vector, Matrix, Tensor

    • Matrix Operations

      • Requirements

      • Addition

      • Multiplication

      • Transposition

      • Inverse

Week 2 - Machine Learning Pipeline

  • First Machine Learning Pipeline - Iris Dataset

    • Load libraries

    • Loading Iris dataset

    • Summarizing the dataset

    • Visualizing the dataset

    • Evaluate a few algorithms

    • Making some predictions

Week 3 - EDA and Intro to Neural Networks

  • Exploratory Data Analysis

    • Statistical Concepts

      • Mean

      • Variance

      • Kurtosis

      • Covariance

    • Data Cleansing

      • Transform Data

      • Treatment of Missing Data

      • Outliers

      • Scaling

      • Correlation

      • Feature Selection

    • Feature Engineering

      • Time Series

      • Dimensionality Reduction

Week 4 - Neural Networks in Python

  • Introduction to Neural Network Structure

    • Simple Neuron structure

    • Simple Neural Network

    • Deep Neural Networks

  • Basic Neural Network in Python

    • No model libraries

    • XOR data set

    • Feed forward

    • Explanation of Back Propagation

    • Matching code

 

  • Same model, apply to Iris data set

Week 5 - Linear Regression

  • Statistical Model Development - Linear Regression

    • Description of Linear Regression

    • Problem Description - Simple Data Set

    • Linear Regression methods

      • Matrix Solution

        • Mathematics derivation

        • Python Code

      • Gradient Descent

        • Development

        • Python Code

    • Comparison to Neural Networks

Week 6 - Classification: Logistic Regression

  • Logistic Regression

    • Model description

    • Python Implementation

  • Classification Evaluation

      • Accuracy/Precision/Recall/F1

      • ROC Curve / AUC

Week 7 - Miscellaneous Modeling

  • Unsupervised Models - K-means Clustering

    • Methodology

    • Example in Python

  • Other Statistical Supervised Models

    • SVM

    • KNN

  • Recommendation Engine

    • Methodology

      • Collaborative Filter

      • KNN

    • Development in Python

Week 8 - Decision Trees, Ensembling

  • Decision Trees

    • Model description

    • Iris example in Excel

    • Development from Scratch

    • Sci-Kit Learn version

  • Ensemble Trees

    • Bagging

    • Random Forest

Week 9 - Decision Trees - Boosting

  • Gradient Boosting

  • Adaboost

  • XGBoost

  • LightGBM

  • CatBoost

Part II - Deeper Dive

Week 10 - Sentiment Analysis with NLP

  • Data Description

  • Data Conversion

  • Stemming

  • Vectorization

    • One-Hot Encoding

    • TF-IDF

    • Word2Vec

    • Other Embeddings

  • Model Fit

  • Model Evaluation

Week 11 - Time Series Modeling

  • RNN/LSTM

  • Language models

  • Sales data

  • Stock Market predicting

Week 12 - Image Recognition

  • Data Sets

    • MNIST

    • ImageNet

  • Traditional Model applications

    • Logistic Regression

    • Random Forest

    • Neural Networks

  • CNN

    • Development

    • Application on MNIST

  • YOLO

    • Description

    • Demo on Video

Week 13 - ML Platforms and Abstractions

  • Platforms

    • AWS

    • Google Cloud

    • Azure

    • NVIDIA/Cuda

    • Spark

  • Abstractions

    • Tensorflow

    • Theano

    • Torch

    • Keras

    • Fast.AI

Week 14 - Reinforcement Learning

  • Architecture

    • Environment

    • Agent/Algorithm

    • States

    • Reward

    • Action

  • ​Q-Learning Algorithm

  • Multi-Armed Bandit Problem

  • Examples​

    • ​Gridwalk

    • Cartpole