Evaluation MLPClassifier Python 

การทำนายข้อมูลด้วย MLPClassifier (Test)

..

👉..

1. # New samples to test

2. # Load data from CSV

3. df_test = pd.read_csv('electric_signal_data_test.csv')

4. features_test_ture = df_test.iloc[:, :-1].values  # Features

5. # Separate features and labels

6. features_test = df_test.iloc[:, :-1].values  # Features

7. labels_test = df_test.iloc[:, -1].values     # Labels

8. 
   

9. # Standardize features

10. scaler_test = StandardScaler()

11. features_test = scaler_test.fit_transform(features_test)

12. 
    

13. # Encode labels

14. label_encoder_test = LabelEncoder()

15. labels_encoded_test = label_encoder_test.fit_transform(labels_test)

16. 
    

17. # Standardize new data

18. new_samples_scaled = scaler_test.transform(features_test)

19. 
    

20. # Predict class

21. predictions = model.predict(features_test)

22. 
    

23. # Print predictions for new samples

24. for i, sample in enumerate(features_test_ture):

25.     print(f"Sample {sample} is predicted to be class {label_encoder_test.inverse_transform([predictions[i]])[0]}")

26.     

27. y_true = labels_encoded_test

28. y_pred = predictions

29. labelss = [2, 1, 0]

30. 
    

31. # Generate confusion matrix

32. cm = confusion_matrix(y_true, y_pred, labels=labelss)

33. 
    

34. # Print confusion matrix

35. print("Confusion Matrix:\n", cm)

36. 
    

37. # Print accuracy

38. accuracy = accuracy_score(y_true, y_pred)

39. print(f"Test Accuracy: {accuracy:.4f}")

40. 
    

41. # Display confusion matrix as a plot

42. disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=labelss)

43. disp.plot(cmap=plt.cm.Blues)

44. 
    

45. # Set the title and show the plot

46. plt.title('Confusion Matrix')

47. plt.tight_layout()

48. plt.show()

เมื่อรันโปรแกรมแล้วผลลัพท์ที่ได้