Module 1: Data Analysis using Numpy and Pandas

1. Numpy

• Numpy Vector and Matrix
• Functions – arange (), zeros (), ones (), linspace (), eye (),
• reshape (), random (), max (), min (),
• argmax (), argmin (), shape and dtype attribute
• Indexing and Selection
• Numpy Operations – Array with Array, Array with Scalars,
• Universal Array Functions

1. Pandas

• Pandas Series
• Pandas Data-Frame
• Missing Data (Imputation)
• Group by Operations
• Merging, Joining and Concatenating Data-Frame.
• Pandas Operations
• Data Input and Output from wide variety of formats like csv, excel, db and html etc.

Module 2: Data Visualization using Matplotlib, Seaborn, Pandas-in built

1. Matplotlib

• plot() using Functional approach
• multi-plot using subplot()
• figure() using OO API Methods
• add_axes(), set_xlabel(), set_ylabel(), set_title() Methods
• Customization – figure size, impoving dpi, Plot ap-pearance,
• Markers, Control over axis appearance and special Plot Types

1. Seaborn

• Distribution Plots using distplot(), jointplot(), pair-plot(), rugplot(), kdeplot()
• Categorical Plots using barplot(), countplot(), box-plot(), violinplot(), stripplot(), swarmplot(), fac-torplot()
• Matrix Plots using heatmap(), clustermap()
• Grid Plots using PairGrid(), FacetGrid()
• Regression Plots using lmplot()
• Styles and Colors customization

1. Pandas Built-in

• Histogram, Area Plot, Bar Plot, Scatter Plot, Box-plot, Hex-plot, Kde-plot, Density Plot e. Choro-pleth Maps
• Interactive World Map and US Map using Plotly and Cufflinks Module

Module 3: Jupyter Notebook

• Introduction
• Basic Commands
• Keyboard Shortcut and Magic Functions

Module 4: GIT

• Distribution Version Control System
• How internally, GIT Manages Version Control on Changesets.
• Creating Repository
• Basic Commands like, git status, git add, git re-move, git branch, git
• checkout, git log, git cat-file, git pull, git push, git commit
• Managing Configuration – System Level, User Lev-el, Repository level

Module 5: Machine Learning Introduction

• What is Machine Learning?
• Machine Learning Process Flow-Diagram
• Different Categories of Machine Leaning –
  • Supervised
  • Unsupervised
  • Reinforcement Learning
  • Scikit-Learn Overview
  • Scikit-Learn cheat-sheet

Module 6: Regression

 Linear Regression

Robust Regression (RANSAC Algorithm)

Exploratory Data Analysis (EDA)

Correlation Analysis and Feature Selection

Performance Evaluation –

Residual Analysis,

Mean Square Error (MSE)

Co-efficient of Determination R^2

Mean Absolute Error (MAE)

Root Mean Square Error (RMSE)

Polynomial Regression

Regularized Regression

Ridge, Lasso

Elastic Net Regression

Bias-Variance Trade-Off

Cross Validation

Hold Out

K-Fold Cross Validation

Data Pre-Processing –

Standardization

Min-Max

Normalization and Binarization

 Module 7: Classification – Logistic Regression

 Sigmoid function

Logistic Regression learning using Stochastic Gradient Descent (SGD)

SGD Classifier

Measuring accuracy using Cross-Validation, Stratified k-fold

Confusion Matrix – True Positive (TP), False Positive (FP), False Negative (FN), True Negative (TN)

Precision, Recall, F1 Score, Precision/Recall Trade- Off, Receiver Operating Characteristics (ROC) Curve.

Gradient Descent

Module 8: Classification –Decision Trees

• CART (Classification and Regression Tree)
• Advantages and Disadvantages and its applications.
• Decision Tree Learning algorithms – ID3, C4.5, C5.0 and CART.
• Gini Impurity, Entropy and Information Gain
• Decision Tree Regression
• Visualizing a Decision Tree using graphviz module.
• Regularization using tuning hyper-parameters using GridSearch CV

 Module 9: Classification – Ensemble Methods

• Bootstrap Aggregating or Bagging
• Random Forest algorithm
• Extremely Randomized (Extra-Trees) Ensemble
• Boosting – AdaBoost (Adaptive Boosting), Gradient Boosting
• Machine (GBM), XGBoost (Extreme Gradient Boosting)

Module 10: Unsupervised Learning – Clustering

• Connectivity- based Clustering using Hierarchical Clustering.
• Ward’s Agglomerative Hierarchical Clustering
• K-Means Clustering
• Elbow Method and Solhouette Analysis

Module 11: Unsupervised Learning – Dimensionality Reduction

• Linear Principal Component Analysis (PCA) reduction.
• Kernel PCA
• Linear Discriminant Analysis (LDA) on Supervised Data

Module 12: Advance NLP with deep-learning overview

• Computational Linguistic.
• History of NLP.
• Why NLP.
• Use of NLP
• NLP Components
• NLP Ambiguity
• Lexical Ambiguity
• Syntactic Ambiguity
• Referential Ambiguity
• Natural Language Understanding
• Natural Language Generation
• Text Planning
• Sentence Planning
• Text Realization

Module 13: Feature Engineering

• Filter Methods
• Wrapper Methods
• Embedded Methods

Module 14: Neural Network

• A Simple Perception
• Neural Network overview and its use case.
• Various Neural Network architect overview.
• Multilayer Network
• Loss Functions.
• The Learning Mechanism.
•  
• Forward and Backward Propagation.
• Gradient Descent
• DL algorithms- ANN, CNN, RNN