SVM分类的准确度相同

Question

我正在使用UCI的汽车评估数据集。我正在尝试使用SVM分类。在模型创建之后，当我使用混淆矩阵计算精度时，即使我改变SVM的参数，每次都获得相同的精度。在下面发布我的代码。

require("e1071");

#Code to read data from csv and convert to numeric
car_data <- read.csv("car.data.csv",header = TRUE,sep = ",",quote = "\"");

#backup original data to other data frame
car_data_bkp <- car_data;

car_data$buying<-as.numeric(car_data$buying);
car_data$maint<-as.numeric(car_data$maint);
car_data$doors<-as.numeric(car_data$doors);
car_data$persons<-as.numeric(car_data$persons);
car_data$lug_boot<-as.numeric(car_data$lug_boot);
car_data$safety<-as.numeric(car_data$safety);
car_data$class<-as.numeric(car_data$class);

#scaling of data
maxs = apply(car_data, MARGIN = 2, max);
mins = apply(car_data, MARGIN = 2, min);

scaled = as.data.frame(scale(car_data, center = mins, scale = maxs - mins));

#sampling of data for train and testing
trainIndex <- sample(1:nrow(scaled), 0.8 * nrow(scaled));
train <- scaled[trainIndex, ];
test <- scaled[-trainIndex, ];

n <- names(train);
f <- as.formula(paste("class ~", paste(n[!n %in% "class"], collapse = " + ")));

svm_model <- svm(formula=f,train,cross = 2,tolerance= 0.00001, cost = 1000,gamma=1);
summary(svm_model);
svm.pred <- predict(svm_model, test[,-7],type = "class");
table(pred = svm.pred, true = test[,7]);
#calculate accuracy
sum(diag(svm.pred))/sum(svm.pred);