如何在SVM中操作多维特征或使用多维特征训练模型？

Question

If I have this input:

"a1,b1,c1,d1;A1,B1,C1,D1;α1,β1,γ1,θ1;Label1"  
"... ... "  
"an,bn,cn,dn;An,Bn,Cn,Dn;αn,βn,γn,θn;Labelx"

Array expression：

[
 [[a1,b1,c1,d1],[A1,B1,C1,D1],[α1,β1,γ1,θ1],[Label1]], 
 ... ... ... ... 
 [[an,bn,cn,dn],[An,Bn,Cn,Dn],[αn,βn,γn,θn],[Labelx]]
                                                     ]

Instance:

[... ... ... ...
 [[58.32,453.65,980.50,540.23],[774.40,428.79,1101.96,719.79],[503.70,624.76,1128.00,1064.26],[1]], 
 [[0,0,0,0],[871.05,478.17,1109.37,698.36],[868.63,647.56,1189.92,1040.80],[1]],
 [[169.34,43.41,324.46,187.96],[50.24,37.84,342.39,515.21],[0,0,0,0],[0]]]

Like this:
There are 3 rectangles,and the label means intersect,contain or some other.
I want to use 3 or N features to train a model by SVM.
And I just learn the " python Iris SVM " code.What should I do?

The Opinion:
this is my try：

from sklearn import svm
import numpy as np
mport matplotlib as mpl
from sklearn.model_selection import train_test_split

def label_type(s):
    it = {b'Situation_1': 0, b'Situation_2': 1, b'Unknown': 2}
    return it[s]


path = 'C:/Users/SEARECLUSE/Desktop/MNIST_DATASET/temp_test.data' 
data = np.loadtxt(path, dtype=list, delimiter=';', converters={3: 
label_type})

x, y = np.split((data), (3,), axis=1)
x = x[:, :3]
x_train, x_test, y_train, y_test = train_test_split(x, y, random_state=1, 
train_size=0.6)

clf = svm.SVC(C=0.8, kernel='rbf', gamma=20, decision_function_shape='ovr')
clf.fit(x_train, y_train.ravel())

Report Error:

Line: clf.fit(x_train, y_train.ravel())

ValueError: could not convert string to float: 

If I try to convert the data:

x, y = np.split(float(data), (3,), axis=1)

Report Error:

Line: x, y = np.split(float(data), (3,), axis=1)

TypeError: only length-1 arrays can be converted to Python scalars

Answer 1

I have few questions before I go for an answer:

Q1. What kind of data you are using to train SVM model. Is it image data? If image data then, is it RGB data? The way you explained you data it seems you are intended to do image classification using SVM. Correct me if I am wrong.

Assumption Let say you have image data. Then please convert to gray scale. Then you try to convert entire data into numpy array. check numpy module to find how to do that.

Once you data become numpy array then you can apply your model.

Let me know if that helps.

Answer 2

SVMs were not initially designed to handle multidimensional data. I suggest you flatten your input features:

x, y = np.split((data), (3,), axis=1)
x = x[:, :3]

# flatten the features
x = np.reshape(x,(len(x),-1))

x_train, x_test, y_train, y_test = train_test_split(x, y, random_state=1, 
train_size=0.6)

clf = svm.SVC(C=0.8, kernel='rbf', gamma=20, decision_function_shape='ovr')
clf.fit(x_train, y_train.ravel())