Description
Artificial Intelligence Generally is based on data. In some cases we observed Machine Learning Models being biased and performing poorly when used. How do you boost the performance and Accuracy of Machine Learning Models. This is what we will learn step by step and as well leverage Techniques such as Log Loss, Cost Function Optimization, Gradient Descent etc.
Abstract
In this practical Codelab session, we will dive into the basics of building Machine Learning Models with an increased performance using the leveraging best practices for techniques such as:
- Feature Engineering
- Log loss
- Gradient Descent
- Vectorization
- Weight
- Cost Function Optimization
- Identifying the Right Metrics for each use case.
Throughout this session, we will use Jupyter Notebooks and Machine Learning Libraries including:
- Scikit Learn
- Numpy
- Pandas
< Content Outline >
- What is Precision Engineering and why we need it.
- Techniques in Precision Engineering
- Feature Engineering
- Hyperparameter Tuning
- Model Selection
- Regularization
- Cross-Validation
- Ensemble Methods
- Feature scaling
< Prerequisites for Attendees >
To attend the Precision Engineering for Machine Learning session, the following prerequisites are recommended:
- Basic Knowledge of Machine Learning
- Programming Skills (preferably Python)
- Mathematics and Statistics Background
- Data Analysis Skills:
- Familiarity with Machine Learning Tools and Frameworks e.g numpy, etc
Please note that these prerequisites will help attendees effectively understand and apply precision engineering techniques in machine learning.
However the goal is to keep it as simple as possible and free for anyone who is interested in Machine Learning.
Come let's dive into the world of Machine Learning.
Audience level: Intermediate or Advanced