Artificial Intelligence for Marketing Industry
ML Review 4 hrs
INTRODUCTION TO DATA SCIENCE 1 hr
What is analytics and what is data science ?
Relevance in the industry
Types of problems and business objectives in various industries
Overview of analytical tools
List of steps in Analytics projects
Why Python for data science ?
PYTHON REVIEW 2 hrs
INTRODUCTION TO DEEP LEARNING
What are limitations of deep learning
What is deep learning
Advantage of Deep Learning over Machine Learning
Real life use cases of Deep Learning
INTRODUCTION TO AI
History of AI
Modern Era of AI
How is this era of AI different ?
Role of ML and DL in AI
Hardware for AI (CPU vs GPU vs TPU)
Software Frameworks for AI
DL Frameworks for AI
Industry Applications of AI
Why Python
Parameters and Hyperparameters
Three flavors of ML
Supervised Learning Fundamentals
Regression or Classification
Training and Evaluating Classification models
Training and Evaluating Regression models
Unsupervised Learning Fundamentals
Principal Component Analysis
Clustering with k
DEEP LEARNING BASICS AND NEURAL NETWORKS
Deep Learning Basics
Train your first neural net
Add the layers to your Net
Guess the architecture
Roll in the deep nets
basic review - 4 hrs
DL - 4 hrs
Linear regression using DL - case study - 2 hrs
Logistic regression using DL - case study - 2 hrs
RNN/GRU/LSTM - 4 hrs
forecasting using RNN - case study - 2 hrs
multi step forecasting using LSTM - case study - 2 hrs
Auto encoder + case study - 4 hrs
recommendation engine + case study - 4 hrs
market basket, clustering, pca, t-sne - 4 hrs
e study - 2 hrs
se study - 2 hrs
M - case study - 2 hrs
study - 4 hrs
t-sne - 4 hrs
Welcome! Course introduction
Introduction to neural networks
Setting up the working environment
Your first machine learning algorithm
TensorFlow - An introduction
4. Introduction to Keras
5. Some Preliminary Tensorflow and Keras Applications
8. Neural Network for Tensorflow & Keras
9. Deep Learning For Tensorflow & Keras
10. Convolution Neural Network (CNN) For Image Analysis
11. Autoencoders With Convolution Neural Networks (CNN)
12. Recurrent Neural Networks (RNN)
Going deeper Introduction to deep neural networks
Backpropagation. A peek into the Mathematics of Optimization
Overfitting
Initialization
Gradient descent and learning rates
Preprocessing
The MNIST example
Business case
–Introduction to Deep Learning 1 hr 1
Revolution in Artifical Intelligence
Limitations of Machine Learning
What is Deep Learning
Advantages of Deep Learning over Machine Learning
Real life use cases of Deep Learning
–Review of Machine Learning and Python
–NumPy and Pandas Crash Course 1 hr 1
Introduction to NumPy
NumPy Arrays
Numpy Index Selection
NumPy Operations
NumPy Exercises
Introduction to Pandas
Pandas Series
Pandas DataFrames - Part One
Pandas DataFrames - Part Two
Pandas Missing Data
GroupBy Operations
Pandas Operations
Data Input and Output
Pandas Exercises
–Visualization Crash Course 1 hr 1
Introduction to Python Visualization
Matplotlib Basics
Seaborn Basics
Data Visualization Exercises
–Machine Learning Concepts Overview 1 hr 1
What is Machine Learning?
Supervised Learning Overview
Overfitting
Evaluating Performance - Classification Error Metrics
Evaluating Performance - Regression Error Metrics
Unsupervised Learning
Case Study 1
–Understanding Neural Networks 4 hrs 4
How Deep Learning works ?
Activation Functions, Hidden Layers, Hidden Units
Perceptron Model
What is Tensorflow
Tensorflow Code basics
Graph Visualization
Tensorflow Constants, Variables and Placeholders
Creating a Model in Tensorflow
Case Study 2
–Deep Dive into Neural Networks with Keras 8 hrs 8
Limitations of a single perceptron
Illustration of Multi layer perceptron
Backpropagation
Cost Functions and Gradient Descent
What is Keras
TensorFlow vs. Keras Explained
Keras Syntax Basics - Part One - Preparing the Data
Keras Syntax Basics - Part Two - Creating and Training the Model
Keras Syntax Basics - Part Three - Model Evaluation
Keras Regression Code Along - Exploratory Data Analysis
Keras Regression Code Along - Exploratory Data Analysis - Continued
Keras Regression Code Along - Data Preprocessing and Creating a Model
Keras Regression Code Along - Model Evaluation and Predictions
Keras Classification Code Along - EDA and Preprocessing
Keras Classification - Dealing with Overfitting and Evaluation
TensorFlow 2.0 Keras Project Options Overview
TensorFlow 2.0 Keras Project Notebook Overview
Keras Project Solutions - Exploratory Data Analysis
Keras Project Solutions - Dealing with Missing Data
Keras Project Solutions - Dealing with Missing Data - Part Two
Keras Project Solutions - Categorical Data
Keras Project Solutions - Data PreProcessing
Keras Project Solutions - Creating and Training a Model
Keras Project Solutions - Model Evaluation
Tensorboard
Case Study 3
–Master Deep Networks 2 hrs 2
Why deep networks ?
Why deep networks give better accuracy ?
Understanding of deep network
Illustration of Forward pass and Backward pass in Deep Networks
Different variants of Gradient Descent
Case Study 4
–Convolutional Neural Networks - CNNs 6 hrs 6
CNN Section Overview
Image Filters and Kernels
Convolutional Layers
Pooling Layers
MNIST Data Set Overview
CNN on MNIST - Part One - The Data
CNN on MNIST - Part Two - Creating and Training the Model
CNN on MNIST - Part Three - Model Evaluation
CNN on CIFAR-10 - Part One - The Data
CNN on CIFAR-10 - Part Two - Evaluating the Model
Downloading Data Set for Real Image Lectures
CNN on Real Image Files - Part One - Reading in the Data
CNN on Real Image Files - Part Two - Data Processing
CNN on Real Image Files - Part Three - Creating the Model
Case Study 5
–Recurrent Neural Networks - RNNs 4 hrs 4
RNN Section Overview
RNN Basic Theory
Vanishing Gradients
LSTMS and GRU
RNN Batches
RNN on a Sine Wave - The Data
RNN on a Sine Wave - Batch Generator
RNN on a Sine Wave - Creating the Model
RNN on a Sine Wave - LSTMs and Forecasting
RNN on a Time Series - Part One
RNN on a Time Series - Part Two
Case Study 6
–Natural Language Processing 4 hrs 4
Introduction to NLP Section
NLP - Part One - The Data
NLP - Part Two - Text Processing
NLP - Part Three - Creating Batches
NLP - Part Four - Creating the Model
NLP - Part Five - Training the Model
NLP - Part Six - Generating Text
Case Study 7
–AutoEncoders 4 hrs 4
Introduction to Autoencoders
Autoencoder Basics
Autoencoder for Dimensionality Reduction
Case Study 8
–Recommendation Engine 4 hrs 4
Introduction to Recommendation engine
Recommendation engine using Collaborative Filtering
Matrix Factorization
Recommendation engine using Matrix Factorization
Evaluation metrics for Recommendation Engine
Case Study 9
–Market Basket Analysis 2 hrs 2
Associate Rule Mining
Apriori Algorithm
Case Study 10
–Review of Machine Learning and Python Case Study 1 Given the google play store apps data, predict the rati
–Understanding Neural Networks Case Study 2 Given the health measures data of patients, create a m
–Deep Dive into Neural Networks with Keras Case Study 3 Given the Airbnb data for accomodations, build a mod
–Master Deep Networks Case Study 4 Given a bank's customers behavior data, create a mod
–Convolutional Neural Networks - CNNs Case Study 5 Given the Fashion MNIST data, create a model to pred
–Recurrent Neural Networks - RNNs Case Study 6 Given a restaurant's reservation and visitation data, y
–Natural Language Processing Case Study 7 Given a Women's clothing e-commerce dataset, you n
–AutoEncoders Case Study 8 Given a bank's customers' transaction data, you have
–Recommendation Engine Case Study 9 Given the movie dataset, create a recommendation e
–Market Basket Analysis Case Study 1 Given the big bazaar's data, you need to predict whic
y store apps data, predict the rating for app (Exploration and Regression). Find the important variables which are responsible for high ratin
asures data of patients, create a model to predict if the patient is diabetic or not.
a for accomodations, build a model to predict prices of the accomodation using deep neural network.
mers behavior data, create a model to predict what new products customers will purchase.
NIST data, create a model to predict the one label of the image out of the ten labels available.
reservation and visitation data, you need to predict the total number of visitors for future dates. This information will help restaurants be
othing e-commerce dataset, you need to predict the sentiment of the reviews.
mers' transaction data, you have to identify the value of transactions for each potential customer for which bank can personalize their ser
aset, create a recommendation engine such that after the user has provided the name of a movie he likes, the engine should be able to se
's data, you need to predict which previously purchased products will be in a user's next order.
h are responsible for high rating.
ation will help restaurants be much more efficient and allow them to focus on creating an enjoyable dining experience for their customers
bank can personalize their services at scale.
he engine should be able to select a list of 5 movies from the database that the user will enjoy.
experience for their customers.
–Review of Machine Learning and Python Case Study 1 Given the google play store apps data, predict the rating for app (Exp
–Understanding Neural Networks Case Study 2 Given the health measures data of patients, create a model to predi
–Deep Dive into Neural Networks with Keras Case Study 3 Given the Airbnb data for accomodations, build a model to predict p
–Master Deep Networks Case Study 4 Given a bank's customers behavior data, create a model to predict w
–Convolutional Neural Networks - CNNs Case Study 5 Given the Fashion MNIST data, create a model to predict the one lab
–Recurrent Neural Networks - RNNs Case Study 6 Given a restaurant's reservation and visitation data, you need to pre
–Natural Language Processing Case Study 7 Given a Women's clothing e-commerce dataset, you need to predict
–AutoEncoders Case Study 8 Given a bank's customers' transaction data, you have to identify the
–Recommendation Engine Case Study 9 Given the movie dataset, create a recommendation engine such tha
–Market Basket Analysis Case Study 1 Given the big bazaar's data, you need to predict which previously pu
dict the rating for app (Exploration and Regression). Find the important variables which are responsible for high rating.
s, create a model to predict if the patient is diabetic or not.
build a model to predict prices of the accomodation using deep neural network.
reate a model to predict what new products customers will purchase.
odel to predict the one label of the image out of the ten labels available.
tion data, you need to predict the total number of visitors for future dates. This information will help restaurants be much more efficient a
taset, you need to predict the sentiment of the reviews.
a, you have to identify the value of transactions for each potential customer for which bank can personalize their services at scale.
mendation engine such that after the user has provided the name of a movie he likes, the engine should be able to select a list of 5 movies
redict which previously purchased products will be in a user's next order.
rants be much more efficient and allow them to focus on creating an enjoyable dining experience for their customers.
their services at scale.
ble to select a list of 5 movies from the database that the user will enjoy.