如何使用python分类器作为SVM来训练从xlsx文件导入的文本..？/

Question

我已使用以下代码导入数据，但问题是如何使用它来实现svm。

from xlrd import open_workbook
 import nltk        
 rb = open_workbook('/Users/tejeshwar/Downloads/Defect-Data.xlsx')    
 sheet = rb.sheet_by_index(0)

 data = ()    
 for row_index in xrange(1, sheet.nrows): #train using 500
     temp,add = (),()
     desc,cat = 0,0 #trial
     for col_index in xrange(sheet.ncols):       
         if col_index==2:
             #print col_index
             desc = sheet.cell(row_index,col_index).value
             #print desc
             #print cellname(row_index,col_index)
             desc = "'" + desc
             #temp +=(desc,)
             #print temp
         elif col_index==5:
             #print col_index
             cat = sheet.cell(row_index,col_index).value
             #print cat
             #print cellname(row_index,col_index)
             cat = "'" + cat + "'"
             add = add + (desc,cat)
             #print (add)
         data = data + (add,)

 print 'done'
 training_data = list(data)`

其中desc是描述而cat是类别

如果给出应该预测的描述或将其归类为类别，我必须以某种方式训练模型