onnx / models

A collection of pre-trained, state-of-the-art models in the ONNX format

http://onnx.ai/models/

Apache License 2.0

7.73k stars 1.38k forks source link

Alexnet preprocess #453

Open M-Tonin opened 3 years ago

M-Tonin commented 3 years ago

Ask a Question

Question

Good night, I have been trying to evaluate alexnet but I can not because the result of the model when I use https://s3.amazonaws.com/model-server/inputs/kitten.jpg is totally wrong. I think the problem is pre process or post process ..... Someone know what preprocess/postprocess use ?

Further information

Relevant Area pre-/post- processing

Is this issue related to a specific model?

All Alexnet models.

wenbingl commented 3 years ago

I didn't see any pre/post process steps for Alexnet model, https://github.com/onnx/models/tree/master/vision/classification/alexnet . Can you share more information about the pre/post process?

mengniwang95 commented 2 years ago

Same question, I found the readme said it just uses CenterCrop and I follow its statement but get only about 33% top1 accuracy. Need preprocess details to reproduce the result.

Apisteftos commented 1 year ago

I have the same problem, is any instructions how to do that? The Top1_Acc with 500 samples was 0.013.

Can someone enlighten us?