Dear Sasikaljarif,

Please note that the model which constructed in this block has 1000 outputs corresponding to number of classes in the Imagenet dataset. If you intend to use this model on your own dataset, it is essential to adjust the number of neurons in the last Dense layer to match the number of classes in your dataset. Additionally, before applying the confusion_matrix function or any evaluation metrics, ensure that your y_true (true labels) and y_pred (predicted labels) are in the correct format, compatible with the number of classes you have specified in the last Dense layer.

Sincerely,