Machine Learning Deployment & Architecture

---

1. What are the skillset required from companies for machine learning

2. Skills evaluations for individual participants

3. Life cycle of a machine learning developer

3.1 Problem Understanding

3.2 Data collection

3.3 Data Wrangling

3.4 Choosing right algorithm

3.5 Building Model

3.6 Model Evaluation

3.7 Model Performance Improvement

3.8 Model Finalisation

3.9 Model Deployment

3.10. Model Documentation

4. Best Practices on Machine Learning

4.1 Scrum Methodology Implementation on Machine Learning

4.2 Agile CRISP Implementation on Machine Learning

5. Building Machine Learning Solution Architecture for Banking Domain Project

5.1 Data Flow Design

5.2 Data Capturing

5.3 Running Machine Learning Engine in AWS using Docker

5.4 Implementations of Jenkins for Machine Learning Deployment

6. Live Project on Machine Learning Practical

6.1 Design Business Case for Machine Learning

6.2 Designing Machine Learning Architecture

6.3 Work allocation

6.4 Data Validation

6.5 Data Understanding Phase

6.6 Domain Understanding Phase

6.7 Feature Extraction Phase

6.8 Environment Study and Algorithm Finalisation

6.9 Model Building

6.10 Model Evaluation & Deployment

6.11 Client Feedback or Stakeholder Feedback on Deployed ML Engine

6.12. Performance Improvement

6.13. Change in Tools for better deployment and better features on improvement

6.14. Machine Learning Model Release

7. How to learn different Machine Learning Tools Faster

7.1 List of tools

7.2 Google Collabs, Google ML Kit

7.3 Apache Mahout, Apache Spark

7.4 ML Deployment Eco-System Docker, Jenkins, Django, Flask

7.5. Documentation and Github Release

---

Phase-4: Interview Preparation Phase

---

• Interview Pattern of Data Science& AI Interviews

• Most frequently asked Interview Questions with Answers & How to explain in interviews

• Most Frequently asked Python Interview Questions & Coding Tests & How to explain in interviews

• Most Frequently asked Statistics Interview Questions & How to explain in interviews

• Most Frequently asked ML/DL Interview Questions & How to explain in Interviews

• Most Frequently asked NLP Interview Questions & How to explain in interviews

• Preparing Data Scientist/AI Resume with Proper Roles & Responsibilities, Analytical skills & live Projects

• How to explain about your team members, project methodology, project lifecycle e.t.c
• How to speak in interviews based on the job description
• How to explain tell me about project which you worked? Live Explanation over the video call to understand confidence and e.t.c

• Interview Etiquette to follow as Data Scientist/AI Engineer

• Resume Marketing, Linkedin Marketing, Keywords optimisation, job profile optimisation e.t.c.

• Mock Interviews to benchmark your skill(Any number of mocks until you feel confident)

---

Phase-3: Advanced Deep Learning, NLU & Reinforcement Learning

---

• Introduction to Biological Neuron vs Artificial Neuron. Working mechanism of Feed Forward Neural Network

• Deep Dive into Optimizers, Activation Functions, Loss Functions and Understanding basics of Tensorflow.

• Back-Propagation & Scripting of Feed Forward Neural Network.

• Feed Forward Neural Network Scripting using Tensorflow & Keras Assignment on Feed Forward Neural Network

• RNN, LSTM, BERT, Attention based Neural Networks, Encoder & Decoder, Bi-Directional RNNs & Project Work on Forecasting(Very Important)

• Transfer Learning on NLP like

• Elmo Models

• BERT Models

• GPT Models

• Albert Models

• Mastering Text Mining, NLP, NLU & NLG

• Word 2 Vector, NER, Co-Reference Resolution, Semantic Network, Topic Modelling, TFIDF, POS Tagging, REGEX for NLP, Text Similarity, Context Understanding using BERT

• Handling Audio Data, Text to Audio, Audio to Text

• End-to-End Deployment of NLP Prototype(Live Project)

• Live Project-1

• Introduction to CNN

• Math behind CNN with Custom CNN Architecture

• Different Pre-Trained Models like VGGNet, ResNet, GoogleNet

• Objection Detection, Intersection over Union

• YOLO

• Object Classification

• Objection Detection Project on OpenCV

• Objection Classification using Pytorch

• Transfer Learning, Different types of CNN Architectures Live Project Demonstration

• Live Project-2

• Project Work

• Introduction to AWS Ecosystem

• Various services in AWS

• Getting Started with Sagemaker

• Understanding S3 bucket, EC2 Instances, Docker containers, e.t.c.

• Performing end-to-end deployment of DL Models into Deployment environment, Monitoring Controlling, Distribution Computing & Parallel Computing

• Understanding the Deep Math Behind RBM,

• Understanding GANS & Different Types of GANs

• Neural Style Transfer with Project

• Variation Autoencoders

• KL Divergence

• Generative Models for Advanced DL Tasking
• Live Project Demonstrations

• Parallel Computing, Distributed Computing, Multi-threading, Multi- Processing, CUDA Architecture, GPU computing, Tf. Fedarated

• Live Project-3

• Different Modes of Model Deployment

• Model Persistance in Keras

• Restoring & loading saved model

• Tensorflow Serving

• Tensorflow Serving Docker

• Deploying on Kubernetes

• Tensorflow Lite

• Introducing ONNx

• Deep into Reinforcement Learning

• Reinforcement Learning

• Reinforcement Learning Framework

• Elements of Reinforcement Learning

• Mathematical Formulation of Reinforcement Learning

• Open AI Gym

---

---