Unexpected Evaluation Scores Training on OpenKBP Dataset

[verse-armour]

Thank you for your excellent work!
I would like to express my sincere appreciation for the high-quality work you have shared. I have diligently followed the instructions provided in your repository and successfully executed the train.py script for 1,600 epochs using the OpenKBP dataset. Subsequently, I ran the pred.py script to evaluate the model's performance.
However, the evaluation results I obtained were as follows:
Dose Score: 6.860165596008301
DVH Score: 3.493969800964223
I am reaching out to see if there might have been any errors or oversights in the way I applied your code. The scores seem unexpected, and I am keen on understanding if there's a step I might have missed or if there's an aspect of the code that could be adjusted for potentially better outcomes.
Your guidance on this matter would be greatly appreciated. Thank you for your time and support.