What you’ll learn
- Prepare and inspect a permitted image dataset without leaking test information.
- Compare a baseline and inference workflow using meaningful quality and resource measures.
Proposed course content
5 sections · 11 illustrative lesson records
Data permissions and preparation3 lessons
- Introduction to Applied AI & Computer Vision
Explore an image-to-model workflow with data permissions, reproducible baselines and transparent evaluation of errors.
Public introduction · Video not supplied - Select shareable or licensed examples
Select shareable or licensed examples. Module: Data permissions and preparation.
Enroll to access · Video not supplied - Review labeling, image quality and dataset splits
Review labeling, image quality and dataset splits. Module: Data permissions and preparation.
Enroll to access · Video not supplied
Image processing foundations2 lessons
- Explore preprocessing and simple visual features
Explore preprocessing and simple visual features. Module: Image processing foundations.
Enroll to access · Video not supplied - Inspect how transformations affect the task
Inspect how transformations affect the task. Module: Image processing foundations.
Enroll to access · Video not supplied
Models and inference2 lessons
- Run a baseline and a permitted pretrained model
Run a baseline and a permitted pretrained model. Module: Models and inference.
Enroll to access · Video not supplied - Record configuration and inference constraints
Record configuration and inference constraints. Module: Models and inference.
Enroll to access · Video not supplied
Evaluation and uncertainty2 lessons
- Measure task-appropriate performance and error patterns
Measure task-appropriate performance and error patterns. Module: Evaluation and uncertainty.
Enroll to access · Video not supplied - Discuss calibration, bias and limitations
Discuss calibration, bias and limitations. Module: Evaluation and uncertainty.
Enroll to access · Video not supplied
Applied prototype2 lessons
- Connect image input to a documented inference workflow
Connect image input to a documented inference workflow. Module: Applied prototype.
Enroll to access · Video not supplied - Report observations without claiming deployment readiness
Report observations without claiming deployment readiness. Module: Applied prototype.
Enroll to access · Video not supplied
Requirements
- Basic Python programming and numerical-array familiarity.
- Introductory statistics and an understanding of training/validation/test splits.
Tools and resources
- A development computer; optional GPU access if the selected model requires it.
- A camera is optional when using permitted recorded images.
- Python, numerical and image-processing tools.
- An appropriate model-inference runtime and version-control tools.
Description
Explore an image-to-model workflow with data permissions, reproducible baselines and transparent evaluation of errors.
Who this course is for
Python learners exploring practical image-processing workflows.
Example practical project
Proposed example: a local image-inspection prototype with a baseline comparison, error analysis and clearly stated limitations.