Topics: The following topics are covered in this module:
- Data Augmentation
- Navigating Hyperparameter Space
- Developing Hyperparameter Intuition for Computer Vision
- Comprehensive Model Evaluation
- Troubleshooting Common Computer Vision Problems
Objectives
By the end of this module, students will be able to:
- Use data augmentation techniques in training.
- Assess the effect of hyperparameter changes on model performance.
- Develop intuition on optimal hyperparameter settings.
- Test and evaluate computer vision models.
- Troubleshoot model issues like overfitting and underfitting.
Watch
Video: Evaluating Models (10:22)
Video: Optimizing Computer Vision Models (7:23)
Topics
Click on the links read the materials within.
- Data Augmentation
- Navigating Hyperparameter Space
- Developing Hyperparameter Intuition for Computer Vision
- Comprehensive Model Evaluation
- Troubleshooting Common Computer Vision Problems
Practice and Apply
Segmentation Exercise
It’s time for our final notebook, where we’ll explore one of the more computationally intensive computer vision tasks: segmentation. As before, you’ll get to train and tinker with a new model. Happy Coding!
The notebooks for the Computer Vision course are located at https://github.com/PracticumAI/computer_vision
Using the same resources as in module 1, complete notebook 03_spot_the_cows.ipynb, which will introduce you to image segmentation.
