Different notebooks from projects.
-
Ensemble_PROJECTS.ipynb (notebook of different ensemble methods (bagging, pasting, boosting and stacking) on the Moon dataset + Mnist Dataset. As Predictors I used models such as Decision Trees (building RandomForests + ExtraForest), Support Vector Machines (LinearSVC)
-
Dimensionality_reduction_PROJECTS.ipynb (notebook with different dimensionality reduction algorithms on the mnist data set and on the 3d to 2d transformation. Principal Component Analysis (PCA), t-Distributed Stochastic Neighbor Embedding (t-SNE), LLE and MDS. Test on building a Pipeline with a reduction and then using a RandomForestClassifier. Combining multiple reductin algorithms (PCA+t-SNE)
-
Unsupervised_learning.ipynb notebook with the Unsupervised learning: clustering with K-Means (finding the optimal number of clusters: inertia, sillhoette score), Gaussian Mixture: anomaly detection.