NSDC (National Skill Development Corporation) - Certificate Program in Artificial Intelligence and Machine Learning 

Python Programming

Installation and Set-upPython Basics - Syntactics,Variable Types, OperatorsFunctions - Built-in, Library and Custom, ArgumentsData Structures and Operations etc

Git and Version Control

Doing version control on local sytem, branching, creating remote repository on github, working collaboratively on github, raising PR, merging PR

Data Analysis using Pandas & Numpy

Introduction to Pandas and Numpy ModulesPandas Basics - Data File Handling, Row/Columns Handling, Slicing, Drop, Sort, New Variable Creation, Observing Frequency CountPandas Advanced

Data Visualization

Importance of Visualizing Data through ChartsTypes of Charts and their Best UsesBasics of Visualization in Python Using Matplotlib and Seaborn:Components of a Plot, Subplots, Functionalities of a PlotPlotting Data Distributions,Univariate distributions

SQL Foundations

Databases - Schema, Table, Relations.Basics of SQL, SQL commands - SELECT, FROM, WHERE, AND, OR, NOT, Pattern matching, sorting, Orderby, Comments and  Operators Group by, aggregate functions

EDA

Doing sanity checks on dataFinding relationship between continous variablesFinding relationship between categorical variables

Prob Dist: Binomial, Poisson

Identifying scenarios where a binomial or a poisson distribution can be usedEnumerating the values that a random variable can takeComputing probability mass using binomial or poisson distribution

Intro to Hypothesis testing

Be able to formulate correct form of null and alternate hypothesis when underlying random variable is binomial or poissonBe able to arrive at the correct formulation of p-valueBe able to use CLT to tackle scenarios involving large sample means

ANOVA

Be able to correctly identify where to use ANOVA, 2 sample t test or chi square test of factor association.Be able to formulate correct null and alternate hypothesis for each of the testsBe able to arrive at a business decision once the test has been done

Predictive Modelling Overview

Undertand the idea of predictor variables and target variables.Outline how a trained model is to be validatedIdentify regression, classification and clustering tasks

Linear Regression

OLS Regression ModelPredicting continuous variableUsing Gradient Descent for Linear Regression, Model evaluation using loss functions, RMSE, R-Square, Stochastic Gradient Descent

Logistic Regression

Predicting a binary variable, interpreting model output, using Python to create a logistic model – using statistics and machine learning methodsChecking model diagnosticsConcept of Confusion Matrix and Computing Accuracy Metrics, ROC, AUC, doing kfold cross validation

Tree Based Ensembles

Decision TreesChecking model diagnosticsBagging and Boosting TechniquesRandom ForestOOBHyperparameter tuning using GridSearch

K-Means Clustering

Introduction to clusteringDistance normsK-means clustering, Elbow method, Silhouette Score, Profiling

Classical NLP

Tfidf featurization and text classificationUsing spacy to handle classical tasks such as lemmatization, pos tagging, dependency parsing Building topic models and discovering key-terms

Deploying Models and Creating a model pipeline

Building a model training pipeline.Building model as an api service

Introduction to Deep Learning

Introduction to Neural Networks: Introduction to Neuron, Activation functions - sigmoid, tanh, relu, etc Loss functions - cross-entropy loss, MSE, etc Optimization Techniques - Gradient Descent, Batch Gradient, Mini-batch Gradient and Stochastic Gradient etc. Building simple MLP using numpy

Deep Learning using Pytorch

Building datasets and dataloadersDefining custom modelsUsing early stopping and logging Using different types of optimizers

Deep learning for NLP Tasks

Word vectors and embedding layersSequential Processing using RNN and LSTM LayersAttention mechanism and encoder-decoder architecture Using hugging face to build bert based models