Ng機器學習 Week9 Anomaly Detection

Anomaly Detection

Q4錯誤

－3，4

Recommender Systems

額是1,2，後來忘記截圖了

Ex

estimateGaussian.m

mu = mean(X);
sigma2 = var(X,opt=1);

selectThreshold.m

predictions = (pval < epsilon);

tPos  = sum((predictions == 1) & (yval == 1));
fPos = sum((predictions == 1) & (yval == 0));
fNeg = sum((predictions == 0) & (yval == 1));

precision = tPos / (tPos + fPos);
recall = tPos / (tPos + fNeg);

F1 = (2 * precision * recall) / (precision + recall);

cofiCostFunc.m

errors = (X*Theta' - Y) .* R;
J = 1/2 * sum(sum(errors .^2)) + lambda/2 * sum(sum(Theta.^2)) + lambda/2 * sum(sum(X.^2));
X_grad = errors * Theta + lambda * X;
Theta_grad = errors' * X + lambda * Theta;