0

Content: Hello. I am attempting a binary classification task for Alzheimer's Disease (AD) and Mild Cognitive Impairment (MCI) using 3D grayscale PET brain images with pytorch.

Data: Data from 282 patients (158 with AD, 124 with MCI) and the respective mmse values for each patient. Models: DenseNet, Graph Convolution Network (GCN) Validation: Nested Stratified 5-Fold Cross Validation The issue I'm facing is that the performance of the GCN model is abnormally higher compared to DenseNet. I suspect there might be data leakage or some problems in my code. If you could review the attached code and point out any issues or areas of concern, I'd greatly appreciate it. I am also prepared to offer compensation if necessary.

I was expecting the debugging code to help verify if each segment of the main code was functioning correctly.The main code didn't behave as intended despite the debugging checks.text https://drive.google.com/file/d/1U5CAHyzV7NXIgy4cumkGkl0MWXQ0O1V3/view?usp=drivesdk

이규빈
  • 1
  • The file you have provided is unavailable (in your google drive trash folder). Can you please provide a functioning link to your code (perhaps using GitHub gists)? – Mooshua Sep 01 '23 at 21:21

0 Answers0