GianlucaVesentini commented on issue #7398: URL: https://github.com/apache/incubator-mxnet/issues/7398#issuecomment-1175218875
I think the problem is that `ImageRecordIter` also performs data augmentation on the images (as explained [here](https://github.com/apache/incubator-mxnet/discussions/12671#discussioncomment-62256)), while if you use `cv2.imread()` you obviously have to do that on your own. I'm facing the same issue: I trained a `mobilenetv3_small` model using a .rec file because it is clearly the preferred way, but now I'm not able to reproduce the same pre-processing operations on images before inference and this gives me completely different results. Is there a solution to this problem? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
