DeepSite - 🧬 RemixImplementation from scratch with MNIST digit classification
Weak learners (depth-1 decision trees) that make predictions based on a single feature threshold.
Sample weights are updated to focus on misclassified examples in each boosting round.
Each stump's contribution is weighted by its accuracy (β = ½ ln((1-err)/err)).
Binary classification task distinguishing between digits 0 and 1.
PCA applied to reduce 784 features to 5 principal components.
1000 samples per class for training, full test set for evaluation.
Training Accuracy
0%
Validation Accuracy
0%
Test Accuracy
0%