An AI Engineer’s technical overview of machine learning life cycle
If you are solving a problem using machine learning, no matter if the problem is a very small business problem or a super large one, the end-to-end machine learning life cycle will be the same.
The machine learning life cycle has 4 stages and these 4 stages are:
1. Data Preparation
2. Model training and Tuning
3. Model deployment and monitor
4. Inference or model serving
The internal steps at each life cycle stage could be more or less however there will always be 4 of the above stages in any machine learning life cycle.
In this video we are going to cover the deep and detailed technical overview of all of these 4 stages as below:
Content Sections:
- Motivation for this video
- Stage 1: Data Preparation
- Data Preparation — Feature Engineering
- Data Preparation — Feature Store
- Data Preparation — Data Artifacts
- Stage 2: Model Training and Tuning
- Stage 3: Model Deployment and Monitoring
- Model Deployment and Monitoring — Online Feature Store
- Model Deployment and Monitoring — MLI
- Stage 4: Inference or Model Serving
That’s all for now.
