Sunteți pe pagina 1din 6

Module 1: Statistics, Linear Algebra & Probability for Machine Learning

 Descriptive Statistics
o Introduction
o Sampling Techniques
o Measures of Central Tendency
o Measures of Variability
o Skewness and Outliers
 Probability
o Introduction
o Probability Laws
o Bayesian Theorem
o Probability Distribution
o Gaussian Distribution
o Sampling Distribution
o Central Limit Theorem
 Normalization
o Z-Score
o Min-Max Method
o Decimal Scaling Method
 Inferential Statistics
o T-Test and Anova
o Chi-Square Test
o Spearman Correlation Coefficient
o Pearson Correlation Coefficient
o Regression Analysis
 Linear Algebra Review
o Review on Matrices
o Operations on Matrices
o Eigen Values and Eigen Vectors
 Dimensionality Reduction (Principal Component Analysis - PCA)

Module 2: Machine Learning

 Intro to Machine Learning & Data Mining


 Machine Learning
 Data Mining
 Supervised & Unsupervised Learning
 Types of Data
 Data Preprocessing
 Frequent Item Sets
 Association Rules & Apriori Algorithm
 Regression
o Linear Regression
o Polynomial Regression
 Bias and Variance
 Base Classifiers
o Logistics Regression
o Decision Tree-based Methods
o K-Nearest Neighbor
o Neural Networks
o Naive Bayes
o Support Vector Machines
o Kernel Trick in SVM
 Clustering
o K-Mean Clustering
o Hierarchical Clustering
o Cluster Evaluation
 Evaluation of Learning Models
o F1-Score
o ROC
o Lift Curves

Module 3: Python for Machine Learning

 Basics of Python
o General Syntax
o Data Types
 Data Structures of Python
o Lists
o Tuples
o Sets
o Dictionaries
 Fundamentals of Python programming
o Functions
o Methods
o Loops
o Conditional Statements
o Classed & Objects
 Libraries of Data Science
o Numpy
o Pandas
o Matplotlib
o Seaborn
o Sklearn

Module 4: Class Projects

 Market Basket Analysis (Apriori)


 Automotive Price Prediction (Linear Regression Algorithm)
 Fraud Detection Prediction (Logistics Regression Algorithm)
 Customer Churn Prediction (Neural Networks Algorithm)
 Customer Segmentation (K-Means Clustering)

Module 5: Assignments

Module 1: Advanced Machine & Deep Learning

 Introduction to Machine Learning


 Linear Regression
 Logistic Regression (Vectorization)
 Neural Networks
 Deep Learning
 Forward & Backward Propagation
 Neural Network Vectorization
 Neural Networks Hyper-Parameters tuning
 Bias, Variance and learning curves
 Types of activation functions
 SVM & Kernel Trick
 Convolutional Neural Network (CNN)
 Classic CNN’s
 CNN Applications
 Text Mining & TF-IDF
 Recommender Systems
 Sequence Modeling
 Recurrent Neural Network (RNN)
 GRU & LSTM
 Word Embedding (NLP)
 Ensemble Learners: Bagging & Boosting

Module 2: Python for Advanced Machine & Deep Learning

Logistic Regression with Neural Networks MNIST

 Numpy Quick Revision


o Introduction
o Vectors and Metrices
o Numpy Basic Operations
 Image Preprocessing
o Loading
o Visualization
o Vectorization
o Normalization
 Single Perception
o Initializing Weights
o Propagation
o Backward Propagation

Single Layer Neural Networks

 Initializing Weights
 Forward Propagation
 Activation Functions
 Cost FUnctions
 Backward Propagation
 Optimization

Hyper Parameters Tuning

 Regularization
o Ridge and Lasso Regression
o Dropouts
 Optimization
o Momentum
o RMS Propagation
o Adam Optimizaer

Keras and CNN

 Keras
o Introduction
o Applying Vanilla Neural Networks
o Parameter Tunning in Kera

 MNIST Data Classification


o Image Preprocessing
o Applying CNN
o CNN Parameter Tunning
 CNN
o Object Detection
o Residual Nets
o Object Classification with Localization

Natural Language Processing (NLP)

 Introduction to NLP
 Data Preprocessing
o Regular Expression (RegEx)
o Stop Words
o Negotiations
 Data Visulaization
 Categorical Variables
 Word Representation
o One Hot Encoding
o Count Vectorization
o TF-IDF Vectorization
o Word Embedding
 Word2Vec
o CBOW
o Skip-Gram
 Embedding Matrix
 Logistic Regression
 Simple Neural Network
 Deep Neural Network
 Recurrent Neural Network (RNN)
 Gated Recurrent Unit (GRU)
 Long-Short Term Memory (LSTM)
 RNN/LSTM using Embedding Matrix

Ensemble Learning

 Voting Classifier
 Bagging
 Boosting

Recommender Systems

 Dealing with Semi-Structured Data Set


 Content-based VS Collaborative Filtering
 Count-Vectorizer and TF-IDF
 Cosine Similarity
 SVD
 Applying Recommendation on Movie Data Set

Video Analytics

 Introduction to OpenCV
 Face Detection

Module 3: Class Projects

 Project 1: Social Media Analytics


 Project 2: Object Detection
 Project 3: Chatbot
 Project 4: GAN
 Project 5: Machine Translators
 Project 6: Stock-Market Prediction
 Project 7: Video Analytics
 Project 8: Recommender Systems

S-ar putea să vă placă și