Confusion Matrix: [[ 84 0 0 0] [ 0 138 0 6] [ 0 0 139 2] [ 0 2 0 1069]],计算准确率、精确率、召回率、f1值

根据混淆矩阵，我们可以计算出以下结果：

- 正确预测的样本数为 84+138+139+1069=1430
- 总样本数为 84+0+0+0+138+6+0+0+139+2+0+2+0+1069=1440

因此，准确率为：

accuracy = (TP + TN) / (TP + FP + TN + FN)
         = (84 + 138 + 139 + 1069) / 1440
         = 0.9931

精确率为：

precision = TP / (TP + FP)
          = 84 / (84 + 0 + 0 + 0)
          = 1.0000 (类别1)
          
          = 138 / (0 + 138 + 0 + 6)
          = 0.9583 (类别2)
          
          = 139 / (0 + 0 + 139 + 2)
          = 0.9857 (类别3)
          
          = 1069 / (0 + 2 + 0 + 1069)
          = 0.9981 (类别4)

召回率为：

recall = TP / (TP + FN)
       = 84 / (84 + 0 + 0 + 0)
       = 1.0000 (类别1)
       
       = 138 / (138 + 2 + 0 + 0)
       = 0.9857 (类别2)
       
       = 139 / (0 + 0 + 139 + 0)
       = 1.0000 (类别3)
       
       = 1069 / (0 + 138 + 0 + 1069)
       = 0.8856 (类别4)

F1 值为：

F1 = 2 * (precision * recall) / (precision + recall)
   = 2 * (1.0000 * 1.0000) / (1.0000 + 1.0000)
   = 1.0000 (类别1)
   
   = 2 * (0.9583 * 0.9857) / (0.9583 + 0.9857)
   = 0.9718 (类别2)
   
   = 2 * (0.9857 * 1.0000) / (0.9857 + 1.0000)
   = 0.9928 (类别3)
   
   = 2 * (0.9981 * 0.8856) / (0.9981 + 0.8856)
   = 0.9386 (类别4)