This is one of the project assigned to me by Computation Intelligence course.
I was asked to use Perceptron, SVM and new Perceptron (uses kernel) models with specific Flight dataset. I used sklearn models for this purpose and analized the result in detail with visual charts and diagram.
For each model I tested each Hyperparameters to obtain the best value for them in order to increase the overal accuracy.
Used models:
- Perceptron
- GridSearchCV
- RepeatedStratifiedKFold
- SVC (with 4 different kernel)
- LinearSVC
- KerneledPerceptron
In order to maximize the accuracy of Perceptron model for non-linear datasets I reimplemeted KerneledPerceptron to use kernels like gussian, polynomial and linear.
Feel free to share your ideas or any other problems. Pull requests are welcomed.
SVM-Perceptron is released under an MIT license. See LICENSE for more information.