Achieving Low Accuracy for Custom Dataset

[junisabot]

Dear Community,

I hope this message finds you well. My name is JK, and I'm have just embarked journey into deep learning for a while. I've been trying to dive deeper into this area myself and have encountered a few challenges along the way. I was wondering if you could spare a few moments to help me out with a question so that I can continue to identify the issue.

Normal Images : 544

Abnormal Images: 40 (plants turning yellowish)

I am achieving quite a low accuracy on my datasets. I am wondering if this datasets I show is a feasible to be use in anomaly detection? What else I might potentially try out to improve the performance?

Config of Patch Core:
config.txt

Log:
log.txt

Thank you so much for your time, and I genuinely appreciate any help you can offer.

Regards,
JK