CNN W 1 Cats and Dogs Workbook Output

JeainnyKim · November 4, 2021, 2:12pm

Could you please help me understand why validation_loss increases to and ends with 1.823 on the final epoch? The validation accuracy seems to increase normally.

maurizioscibilia · November 4, 2021, 2:32pm

Hi @JeainnyKim ,

loss and accuracy are not always inversely correlated. Loss depends on the difference between raw prediction and class, while accuracy measures the difference between the prediction converted to 0 or 1 (the thresholded prediction) and the class. Accuracy and loss are still inversely correlated most of the times, but it can happen that they are not, like probably in your case.

Best

Topic		Replies	Views
Validation vs Training Graph Convolutional Neural Networks in TensorFlow week-4	1	578	April 15, 2022
Validation Accuracy and Lost better than Training Acc/Loss Convolutional Neural Networks in TensorFlow week-4	2	554	September 6, 2022
Increasing loss for validation set Convolutional Neural Networks in TensorFlow week-1	2	517	June 6, 2022
C2W1 How can accuracy of the validation set improve while the loss is constant Convolutional Neural Networks in TensorFlow week-1	4	326	November 17, 2023
Help in interpreting validation accuracy Convolutional Neural Networks in TensorFlow week-1	2	506	December 8, 2022

CNN W 1 Cats and Dogs Workbook Output

Related topics