Training For AI Professionals : Practical AI
Learn from industry experts, IITians, PhDs (Doctors)
Proposed Duration
-
Five Weekends
Each Weekend
-
Two days (6 hours each day)
Start
-
23rd March 2019 Saturday
Multiple Hands On Workshop
You Can Learn From IITians
The proposed training is for working professionals who need to understand the nuts and bolts of building AI/Deeplearning applications from a practical perspective. This will be done in a hands-on mode combined with an in-depth conceptual rendering of relevant concepts.
The proposed training is for working professionals who need to understand the nuts and bolts of building AI/Deeplearning applications from a practical perspective. This will be done in a hands-on mode combined with an in-depth conceptual rendering of relevant concepts.
You will eventually deploy your Tensorflow based deep learning models on industry leading platforms such as AmazonSageMaker, Google Cloud Platform (GCP) ML Engine and Azure AI.
While several variants of AI trainings exist in the market, this training is intended to represent a complete practical hands-on oriented approach to equip developers with knowhow to build AI applications for your company – right from modelling to production deployment.
-
- Brief Introduction to AI
- What is AI
- Use cases
- Tech Stack
- AI vs ML vs DL
- Setting up the development Environment
- Anaconda
- Jupyter notebooks
- Refresher on Python
- Intro to numpy and Pandas
- In class coding assignment
- Brief Introduction to AI
-
- Basics of ML
- Unsupervised, Supervised and Reinforcement
- Unsupervised ML at a glance
- Clustering, Recommendation Systems
- Code walkthroughs using SkLearn
- Supervised ML at a glance
- Classification (Logistic Regression, Decision Trees)
- Linear Regression
- Code walkthroughs using SKLearn
- In class coding assignment
- Basics of ML
-
- Basics of Deep Learning
- Introduction to Neural networks
- Coding a simple neural network in Keras/Tensorflow
- Deep dive on Training, Loss functions, gradient descent and back Propagation
- Strategies to handle Overfitting and Underfitting
- In class coding assignment( Auto-encoders)
- Basics of Deep Learning
-
- Computer vision and Deep Learning
- Introduction to image processing
- Use cases
- Convolutional Neural Networks (CNN)
- Introduction
- Using OpenCV framework
- Walkthrough of an image processing example using CNN
- In class coding assignment on CNN
- Natural Language Processing (NLP) and Deep Learning
- Introduction
- Use cases
- Recurrent Neural Networks (RNN)
- Introduction
- Walkthrough of an NLP use case using RNN
- In class coding assignment on RNN (Sentiment Analysis)
- Computer vision and Deep Learning
-
- Practical Considerations of Machine Learning
- Overfitting vs underfitting
- Weight Initialization
- Early stopping
- hyperparameter tuning
- Normalization
- Dropouts
- Dataset design and understanding biases in data
- Training on GPUs vs CPUs
- Conversational AI
- Introduction
- Walkthrough on chatbot code example
- In class coding assignment – write your own chatbot
- Emerging areas
- Attention networks
- Generative Adversarial Networks (GANs)
- Introduction to Amazon SageMaker
- Setting up Amazon Account
- Development to Deployment workflow
- Walkthrough of a sample model deployment
- In class coding project
- Practical Considerations of Machine Learning