Pengfight / proj3

The homework3 of data mining

Decision Trees model

Learning decision trees, entropy as criteria, max depth allowed 11

python -m hw3.models.runModels --dataset connect-4 --train_test_split 0.2 --model dtree --seed 12 --max_depth 11 --splitting_criteria entropy

Learning decision trees, Gini as criteria, max depth allowed 11

python -m hw3.models.runModels --dataset connect-4 --train_test_split 0.2 --model dtree --seed 12 --max_depth 11 --splitting_criteria gini

Deep Neural Network

Learning neural network, learning rate of 0.05, batch size of 1 and logging accuracy and loss at every 5 epochs for 50 epochs with seed of 12

python -m hw3.models.runModels --dataset connect-4 --train_test_split 0.2 --model nnet --log_interval 5 --batch_size 1 --epochs 50 --lr 0.05 --seed 12

SVM

python -m hw3.models.runModels --dataset connect-4 --model svm

Logistic Regression

python -m hw3.models.runModels --dataset connect-4 --model lr

Random Forest

python -m hw3.models.runModels --dataset connect-4 --model rf