- 22 (Registered)
When: 4th August 2020 to 8th August 2020
Course Description
This course will introduce the learner to applied machine learning, focusing more on the techniques and methods than on the statistics behind these methods. The course will start with a discussion of how machine learning is different than descriptive statistics, and introduce the scikit learn toolkit through a tutorial. The issue of dimensionality of data will be discussed, and the task of clustering data, as well as evaluating those clusters, will be tackled. Supervised approaches for creating predictive models will be described, and learners will be able to apply the scikit learn predictive modelling methods while understanding process issues related to data generalizability (e.g. cross validation, overfitting). The course will end with a look at more advanced techniques, such as building ensembles, and practical limitations of predictive models. By the end of this course, students will be able to identify the difference between a supervised (classification) and unsupervised (clustering) technique, identify which technique they need to apply for a particular dataset and need, engineer features to meet that need, and write python code to carry out an analysis.
Course Content
-
Day 01 - Module-1 Introduction to ML 0/8
-
Artificial Intelligence & Machine Learning IntroductionLecture1.1
-
Who uses AI?Lecture1.2
-
AI and ML for telecomLecture1.3
-
Supervised & Unsupervised LearningLecture1.4
-
Regression & Classification ProblemsLecture1.5
-
What makes a Machine Learning Expert?Lecture1.6
-
What to learn to become a Machine Learning Developer?Lecture1.7
-
Overview of Machine Learning AlgorithmsLecture1.8
-
-
Module-2 Introduction to Python 0/4
-
Basic syntaxLecture2.1
-
Data structures in PythonLecture2.2
-
FunctionsLecture2.3
-
Object oriented ProgrammingLecture2.4
-
-
Module-3 Data Structure & Data Manipulation 0/9
-
Indexing, Data Processing using Numpy Arrays and PandasLecture3.1
-
Mathematical computing basicsLecture3.2
-
Basic statisticsLecture3.3
-
File Input and OutputLecture3.4
-
Getting Started with DataframesLecture3.5
-
Data Acquisition (Import & Export)Lecture3.6
-
Selection and FilteringLecture3.7
-
Combining and Merging Data FramesLecture3.8
-
Removing Duplicates & String ManipulationLecture3.9
-
-
Module-4 Data Visualization 0/6
-
Data pre processingLecture4.1
-
Data cleaning and data transformationLecture4.2
-
Feature extractionLecture4.3
-
Recursive feature eliminationLecture4.4
-
Feature transformationLecture4.5
-
Discussion on mining massive data setsLecture4.6
-
-
Day 02 - Module-1 Linear Regression - Case Study & Project 0/7
-
Regression Problem AnalysisLecture5.1
-
Mathematical modelling of Regression ModelLecture5.2
-
Gradient Descent AlgorithmLecture5.3
-
Use casesLecture5.4
-
Regression TableLecture5.5
-
Building simple Univariate Linear Regression ModelLecture5.6
-
Multivariate Regression ModelLecture5.7
-
-
Module-2 Linear Regression - 2 0/3
-
Polynomial Linear RegressionLecture6.1
-
Ridge RegressionLecture6.2
-
Lasso RegressionLecture6.3
-
-
Module-3 Logistic Regression 0/13
-
AssumptionsLecture7.1
-
Reason for the Logit TransformLecture7.2
-
Logit TransformationLecture7.3
-
HypothesisLecture7.4
-
Variable and Model SignificanceLecture7.5
-
Maximum Likelihood ConceptLecture7.6
-
Log Odds and InterpretationLecture7.7
-
Null Vs Residual DevianceLecture7.8
-
Chi Square TestLecture7.9
-
ROC CurveLecture7.10
-
Model SpecificationLecture7.11
-
Confusion MatrixLecture7.12
-
Accuracy, Recall, Precision and F1 ScoreLecture7.13
-
-
Day 03 - Module-1 Decision Trees 0/4
-
Forming a Decision TreeLecture8.1
-
Components of Decision TreeLecture8.2
-
Mathematics of Decision TreeLecture8.3
-
Decision Tree Evaluation for use casesLecture8.4
-
-
Module-2 Random Forests 0/2
-
Random Forest MathematicsLecture9.1
-
Examples & use cases using Random ForestsLecture9.2
-
-
Module-3 KNN 0/3
-
KNN TheoryLecture10.1
-
KNN with PythonLecture10.2
-
KNN Exercise with real dataLecture10.3
-
-
Module-4 Bagging and Boosting 0/5
-
Bias Variance TradeoffsLecture11.1
-
BaggingLecture11.2
-
BoostingLecture11.3
-
BootstrappingLecture11.4
-
Ensemble models with real world dataLecture11.5
-
-
Day 04 - Module-1 Naïve Bayes Algorithm 0/8
-
Bayesian TheoremLecture12.1
-
Probabilities – The Prior and Posterior ProbabilitiesLecture12.2
-
Conditional and Joint Probabilities NotionLecture12.3
-
Traditional Approach – Extract Important Features 0 question10 minQuiz12.1
-
Naive Approach – Independence of Features Assumption 0 question10 minQuiz12.2
-
Data Processing – Discretization of Features 0 question10 minQuiz12.3
-
Advantages & Disadvantages of Naïve Bayes Models 0 question10 minQuiz12.4
-
Examples, Projects & Case Studies 0 question10 minQuiz12.5
-
-
Module-2 Support Vector Machine 0/4
-
Concept and Working PrincipleLecture13.1
-
Mathematical ModellingLecture13.2
-
The Kernel Method and Nonlinear HyperplanesLecture13.3
-
Use Cases & Programming SVM using PythonLecture13.4
-
-
Module-3 Recommendation Systems 0/2
-
Market basket AnalysisLecture14.1
-
Collaborative FilteringLecture14.2
-
-
Day 05 - Module-1 Clustering 0/5
-
Hierarchical ClusteringLecture15.1
-
K Means ClusteringLecture15.2
-
Use Cases for K Means ClusteringLecture15.3
-
Programming for K Means using PythonLecture15.4
-
Cluster Size Optimization vs Definition OptimizationLecture15.5
-
-
Module-2 Time Series 0/9
-
Definition of Time SeriesLecture16.1
-
Time Series DecompositionLecture16.2
-
Simple Moving Average MethodLecture16.3
-
Weighted Moving Average MethodLecture16.4
-
Single Exponential Smoothing MethodLecture16.5
-
Double Exponential Smoothing MethodLecture16.6
-
Triple Exponential Smoothing MethodLecture16.7
-
Stationarity of DataLecture16.8
-
ARIMA ModelsLecture16.9
-
-
Module-3 Reinforcement Learning 0/3
-
Upper Confidence BoundsLecture17.1
-
Thompson SamplingLecture17.2
-
Example and use casesLecture17.3
-
-
Day 06 - Module-1 Artificial Neural Networks 0/8
-
Activation Functions for Neural NetworksLecture18.1
-
Optimization Techniques – SGD, ADAM, LBFGSLecture18.2
-
RegularizationLecture18.3
-
Momentum in Neural NetworksLecture18.4
-
Neural Network Tuning and Performance OptimizationLecture18.5
-
Perceptrons, Multilayer Perceptrons, Weights, Biases,Backpropagation,OptimizationLecture18.6
-
Training, Validation and TestingLecture18.7
-
End-to-end development of a multiclass classification problem in Python/MATLABLecture18.8
-
-
Module-2 Convolutional Neural Networks 0/7
-
CNN ArchitectureLecture19.1
-
Introduction to Deep Learning, History, ApplicationsLecture19.2
-
Shallow Networks and Deep NetworksLecture19.3
-
Convolutional Neural Networks, Pooling layers, Fully Connected layers, Activation functions, ReLU, Dropout, loss function and Optimization with respect to CNNLecture19.4
-
Introduction to frameworks Tensorflow, KerasLecture19.5
-
Image classification using convolutional neural networks, visualisation of featuresLecture19.6
-
MNIST Data – Digit Classification using CNNLecture19.7
-
-
Module-3 Recurrent and Recursive Nets 0/4
-
Basic concepts of RNNLecture20.1
-
The Vanishing Gradient ProblemLecture20.2
-
LSTM NetworksLecture20.3
-
Recursive Neural NetworksLecture20.4
-
-
Module-4 Machine learning at scale 0/3
-
AWS ML OfferingsLecture21.1
-
Google mL OfferingsLecture21.2
-
On- Premised DeploymentsLecture21.3
-
Instructor
Arun Panayappan is the Founder of Techcovery Solutions which has been in the training and consulting business for three years now. Arun Panayappan was the Vice President, Platform Engineering at www.goibibo.com which is India’s largest online travel agent. He has over 23 years of IT experience including 4 years of overseas experience.Techcovery has trained over 1000 participants in Data science and mentored 4 start ups during its existence. He has worked as a consulting CTO for companies like cogknit semantics, FirstIPO and Trip38. He is frequently consulted for data science architecture and strategy by corporations across the industry. In his role at Amazon, his technical decision-making skills enabled Kindle Topaz Operations Console to go global and scale seamlessly to meet seasonal load. He drove initiatives to hire and develop the best for the Category Expansions team that doubled in size during his leadership tenure while launching 70 new stores for www.amazon.com every year. He built highly scalable web crawling systems to monitor price changes at competitor sites for millions of products sold by Amazon. This system used statistical machine learning and modeled crawl frequencies using poison distribution. His team built the on-boarding experience of vendor platform(https://vendorexpress.amazon.com) to onboard and manage small and medium size vendors to increase selection and buying options. Prior to joining Amazon, he worked for Kuliza (www.kuliza.com), young start up by ex-Amazon employees, and had served the Research team at Accenture Technology Labs., India. He is an innovative and a tech-savvy leader. He sets a high bar for himself and his team on software design, quality and processes. He has a pleasing personality and excellent communication skills.
Related Courses
