Overview of ML, AI With Python Training
Machine Learning is a global phenomenon, it is the application of artificial giving systems the ability to automatically learn and improve the experience without having to program aspects explicitly. Python considered the preferred language for Machine Learning is used extensively in this course to cover various concepts and drive participants through the depths and various use cases of Machine Learning.
At the end of this training program, participants will have a depth of understanding of
- Python for Machine Learning
- Various Machine Learning concepts & algorithms
- Working with Data for Machine Learning
Duration
3 Days
Prerequisite for ML, AI With Python Training
- Must know /covered Matrix, Linear Algebra and Probability in School.
- This will be conceptual discussion with some code demo (in Python). Python will not be covered.
Course Outline for ML, AI With Python Training
ML, DS, DL, NLP(8 hours)
- What is ML, DS, AI, NLP, DL? Demystifying these terms
- Use cases of ML
- Use cases of processes that are NOT ML
- Supervised ML, Unsupervised ML
- Examples of Supervised ML, Unsupervised ML
- Videos of ML
- Regression and Classification
- What is Data Science? Detailed steps in DS
- Most important step in ML, DS – DATA
- Define Problem and success measures
- Do a Pilot to solve above (This is process point)
- Understand Benchmarks and SOTA
- EDA – Understand Data, Clean Data, Choose fields
- CRISP-DM Process for project execution
- Getting Data based on different scenarios (Data Rich, public, outsource and create own data by designing experiments and data augmentation)
- Regression and Classification Algorithms
- Types of data – quantitative and Qualitative and challenges with those data
- KNN algorithm
- What is KNN
- Challenges with KNN
- When to avoid using KNN
- Intro to SKLearn
- When it can be used
- Embedded Algorithm
Naïve Bayes and Decision Tree(5 hours)
- Naïve Bayes
- Brief Summary of Algo
- NB and Text Analytics
- NB and quantitative data
- Decision Tree
- What is Decision Tree and why it is popular
- Get the rules and insights easily from Decision Tree
- Issues with Decision Tree
- How to address them using advance algorithms like RF, and Boosting
- AI Explainablilty and Interpretable models
Linear and Logistic Regression(4 hours)
- Linear Regression
- How does it help
- Easy interpretable results and importance of each features
- Model significance using stats model. Will show how to interpret the results (with getting into technique)
- Logistic Regression
- How does it help
- Easy interpretable results and importance of each features
- AI Explainablilty and Interpretable models
- Model significance using stats model. Will show how to interpret the results (with getting into technique)
Unsupervised Learning and DL(4 hours)
- Unsupervised Learning
- Learn insights from unsupervised methods, Dimensionality Reduction
- Deep Learning Intro
- Summary of DL models and advantages and disadvantages and some use cases of where we could use them