AI & Machine Learning

Register Now

Day 1
 
Artificial Intelligence Introduction and Environment Setup:

• Introduction to Artificial Intelligence and related Industry Use Cases.
• Basics of Data Science, Machine Learning, Deep Learning and NLP.
• Setting up Anaconda & Python Notebooks.
• Working on notebooks for Machine Learning.
• Setting up Deep Learning Tensorflow Environment.

 
Python Basics:

• Python Basic Data Types.
• Data Structures (List, Tuple, Dictionary).
• Creating, accessing, and slicing tuples, lists and Dictionaries.
• Functions, Control Flow.
• Module and Packages.
• Errors and Exceptions.

 
Python Mathematical Computation Libraries (Numpy) and Scientific (SciPY):

• Numpy Overview.
• Properties and types of Ndim Arrays.
• Accessing Array elements: Indexing, Slicing, Iteration, Indexing with Boolean Arrays.
• Basic operation on NDim Arrays.
• Shape Manipulation.
• Numpy Mathematical operations examples.
• Scipy Overview.
• Scipy Statistical sub package (T-Test, p-value, Anova etc).
• Scipy Linear algebra computation.

 
Data Manipulation and Exploratory Analysis (Pandas):

• Introduction to Pandas.
• Pandas Series.
• Pandas Data Frame.
• Pandas File Read and Write Support.
• Data Operation, Summarization, Slicing & Dicing.
• Data Cleaning Missing Values, Outliers.
• Data Filtering.

 
Data Visualization in Python (Matplotlib):

• Introduction to Data Visualization.
• Matplotlib Features.
• Bar Plot, Line Chart, Box Plot, Scatter Plot.
• Set Axis, Labels, and Legend Properties.
• Controlling axis labels and colors.

 
Day 2
Supervised Machine Learning-1 (Scikit Learn):
(Industry Use case Implementation with Model Tuning)

• Introduction Machine Learning Life Cycle.
• Decision Tree Algorithms.
• Classification Tree.
• Regression Tree.
• Tree based Ensemble Learning.
• Random Forest.
• Gradient Boosting Trees.

 
Model Evaluation, Improvements & Performance Metrics:
(Machine Learning lifecycle for Model Tuning)

• Data Split Practices.
• Cross Validation.
• K-Fold Validation.
• Confusion Matrix.
• ROC Curves.
• Mean Absolute/Square Errors & R-Square.
• Ensemble Learning.
• Model Selection and Finalization.
• Grid Search.

 
Day 3

Supervised Machine Learning-2:

• Support Vector Machine (SVM Model).
• SVM Kernel Training.
• SVM Model hyper parameter Tuning.
• K Nearest Neighbor (KNN Model).
• KNN Model Training.
• Naive Bayes Model.

 
Regression Learning (Industry Use case Implementation with Model Tuning):

• Linear Regression.
• Logistic Regression.
• Regularization.
• Multiple Variables.
• Gradient Descent.
• Model Training and Validation.
• Model Hyper parameter tuning.

 
Forecasting (Time Series Modelling):

• Trend and Seasonal Analysis.
• Different Smoothing Techniques.
• ARIMA Modelling.
• Diff Data Transformation Techniques .
• Model Tuning.

Introduction to Tensorflow and Keras (Deep Learning):

• Tensorflow overview.
• The Programming Model.
• Keras Overview.
• Keras Pre-processing APIs.
• Keras Data Sets.
• Data Pre-processing and optimization.
• Feature Standardization.
• Activations.
• Keras Model Layers.

 
Artificial Neural Network Modelling and Tuning:
(Industry Use case Implementation with Modelling)

• Multi-Layer Perceptron Model.
• Cost Function Formation.
• Training MLP Model.
• Weight initialization.
• Back propagation and Gradient Descent Algorithm.
• Optimizers (ADAM, AdaGrad etc).
• Hyper parameter Tuning.
• Cross Entropy and Soft Max for multi classification.
• Bias Variance tradeoff.
• Non-Linear Activation Function.
• Industry Use case implementation.

 
Convolutional Neural Networks and Recurrent Neural Network:
(Industry Use case Image recognition Implementation with Modelling)

• CNN Architecture.
• Convolution Function.
• Pooling and Flattening.
• Computer Vision.
• Image Acquisition and manipulation.
• Edge Detection, Corner Detection.
• Image Scaling.
• Dropout layer and Regularization.
• Rectifier Linear Unit (Relu).
• Recurrent Neural Network.
• LSTM Time Series Modelling.
• Industry Use Case Implementation.

 
Natural Language Programming (NLP) Overview:

• NLP Introduction.
• NLP role in AI and related use cases overview.
• Python NLP Packages NLTK, Spacy, Gensim etc.

 
NLP Text Pre-processing and Cleaning:

• Steps required to clean the text data.
• Text Pre-processing and cleaning.
• Text Tokenization (word, sentence etc.).
• Object Standardization.
• Noise Entity Removal.
• Removing Stop Words and Punctuations.
• Regular Expression.

 
Text Feature Engineering (Industry Use Case Implementation):

• Different steps of text feature engineering.
• Syntactical Feature Analysis.
• Semantic Analysis.
• Statistical features (TF-IDF).
• Word Embedding.
• Word Vector Analysis.

 
Text Classification Use Case (Industry Use Case Implementation):

• End-to-End implementation of Text Classification Use Case using NLP modelling in Python.
• Feature transformation (word embedding, TF-IDF etc.).
• Model Training and Validation.
• Model Tuning and Evaluation.
• Model Finalization.

 
Industry Use Case Implementation:

• Demand Forecasting: Predict the future Sales of a Product using Multivariate Solutions. (Machine Learning and Deep Learning based Modelling).
• Price Sensitivity Modelling: Find out the optimum price during pricing decision for a product, given the current state of market, the competition and the period of year.
• Inventory Forecasting: Prediction of Stockouts to understand when to restock and how many units to order.
• Warehouse Management: Visual Data based Defect Detection (Quality issues prediction) using AI model (Computer Vision, Image Recognition Modelling).
• Distribution: Prediction of damages to the material due to Transport conditions (IOT sensor data).