Plugin: fine tuning of existing vision models #89
Description
current plugins namely : facematch , deepfake , age-detector all three of them could be benfit from fine tuning . for example, focus on age 5 to 19 images
all of these use onnx models , can a common code approach work for fine tuning ?
Fine-tuning with LoRA adapters in ONNX Runtime
if GPU is required can we use AVD /Unity host for this ?