In this project, one can learn the functioning and intuition behind a powerful class of supervised linear models known as support vector machines (SVMs). By the end of this project, one will be able to apply SVMs using scikit-learn and Python classification tasks, including building a simple facial recognition model.
The project starts with the theory discussing how SVMs work in classfiying the data along with the underlying mathematics of the classifier. Further we plot margins of support vectors from the data and see the functioning of SVMs in practice. After getting a general idea of the model we take up the example of face-recoognition for classification. We have used Grid Search technique and PCA to obtain optimal model and then calculated the model classification report from scikit metrics. Further the confusion matrix is used to see how well our model performed on the dataset.
- Faces of famous politicians of USA :sklearn.datasets