Added check for intergrity and accuracy of the .onnx model

GeekyYouthsInfo / Deep-Learning-Cataract-Detection

Academic: Deep learning-based mobile app to accurately identify the presence and severity of cataracts, enabling early diagnosis and treatment.

0 stars 1 forks source link

Added check for intergrity and accuracy of the .onnx model #19

Closed jona42-ui closed 2 months ago

jona42-ui commented 2 months ago

So, after converting the .pth Pytorch model to .onnx model, I needed to check if the accuracy is still maintained I have added test dataset test the accuracy.

Generally, I verified if the ONNX model exhibits the same accuracy as the PyTorch model by running inferences using the ONNX model and comparing the results.

Comparing with the PyTorch model(final_model.pth)