Unit 1 : (1 week )

Unit 1 : Fundamentals for Data science : (1 week )

CRISP-DM methodology, ETL

SAI Application and use cases

Programming for ML ( python)

Data sourcing, Wrangling and Munging

Python : Numpy, Pandas

Sql basics for data science

Unit 2 : (1 week )

Unit 2 : Exploratory Data Analysis ( EDA) : (1 week )

Qualitative and quantitative data

Measure of central tendency

Covariance, correlation chi square techniques

Data Visualization ( practical in python Matplotlib, pyplot )

Unit 3 : (1 week )

Unit 3 : Statistics foundations for Data Science (1 week )

Contingency table

Probability distribution

Sampling distribution

Confidence Intervals, Hypothesis testing

Bayesian Methods

Practical (Statistics in Python )

Unit 4 : (1 week )

Unit 4: Regression analysis (1 week )

Linear and Multivariate regression

Residual Analysis ( R square goodness of fit)

Identifying Significant features, Feature reduction using AIC

Regularization methods ( lasso, Ridge )

Generalized Linear models

Unit 5 : (1 week )

Unit 5: Machine learning foundations (1 week )

Ml techniques

Variance and Bias

Validation techniques

Dimensionality reduction

Feature engineering, Unbalanced Data Treatment

Unit 6 : (1 week )

Unit 6: Supervised Learning (1 week )

Naive Bayes Classifier

Logistic regression

K-nearest Neighbours

Support vector machine

Association rule mining ( apriori algorithm)

Python Practical - Sk learn module

Unit 7 : ( 3- 4 days )

Unit 7: Decision Trees and ensemble ( 3- 4 days )

Decision Tree

Bagging and boosting ( Random forest, GBM, XG boost)

Python Practical - Sk learn module

Unit 8 : ( 3 days + 1 week )

Unit 8 : Unsupervised learning ( 3 days + 1 week )

Distance methods

Clustering methods

K-medoids, K- modes

Python Practical - Sk learn module

Unit 10 : (1.5 weeks )

Unit 10 : Neural Network basics (1.5 weeks )

ANN application areas

Gradient Descend, Perceptron, MLP, FFN, Back propagation

Regularization - Dropout and batch normalization

ANN for structured Data

LSTM ( long short term memory)

Python practical ( Tensor flow module )

Unit 11 : (1 week )

Unit 11 : Deep learning (1 week )

Convolutional Neural Networks ( CNN’s)

Recurrent Neural Network (ANN’s)

Auto Encoders

Object detection, Image classification, Face detection

Unit 12: RL ( 1.5 weeks )

Reinforcement Learning

Overview of reinforcement learning: the agent environment framework, successes of reinforcement learning

Bandit problems and online learning

Markov decision processes

Monte Carlo learning

Case studies: successful examples of RL systems ( using python - open AI gym module