def dsquared_loss(y_pred, y_train): m = y_pred.shape[0] grad_y = y_pred - util.onehot(y_train) grad_y /= m return grad_y
def squared_loss(model, y_pred, y_train, lam=1e-3): m = y_pred.shape[0] data_loss = 0.5 * np.sum((util.onehot(y_train) - y_pred)**2) / m reg_loss = regularization(model, reg_type='l2', lam=lam) return data_loss + reg_loss