-1

I have built svm kernel polynomial algorithm for predicting health diagnose with 2 labels and 37000+ data in dataset. Training and testing accuracy score is up 0.97. and when i test new data with this agorithm, the result is also as good as the accuracy above. But the problem is when i test this new data via postman, the predictive diagnose went wrong up to 50 percent, the system cannot predict in a good result. What is the problem and what is the solution?

I have tried to test 200 new datas with this algorithm, the accuracy score in google colab is good but on postman is not good. I expected if the predictive model can classify/diagnose in a perfect way ko

molbdnilo
  • 64,751
  • 3
  • 43
  • 82
Wardatus Sholihah
  • 1
  • Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. – Community Aug 05 '23 at 20:57

0 Answers0