mozga-intel edited a comment on pull request #20430: URL: https://github.com/apache/incubator-mxnet/pull/20430#issuecomment-893503256
We need to figure out how to remove some duplications i.e: `QuantizeFcShiftedQuantization` `FcFcShiftedQuantization` [(here](https://github.com/apache/incubator-mxnet/pull/20430/files#diff-93eeb14143daedfc06e4bf9be25dd477cd13ad536c124e45bed92e1787101854R139)). Calculate a position of particular input in the input vector could be more error-proofing. It could refer to the implementation of a more specific structure that gives an ability to set up those attributes. [here](https://github.com/apache/incubator-mxnet/pull/20430/files#diff-fe163642508b98c6d809888be8dad197d653e7b602a639846751a7ac1bcc64c8R111). In this way we can lack errors and unexpected mistakes. Last questions: - Could you please tell me which model uses that? - Does the following patten is disabled by default, If the answer is 'yes', will we have a plan to enable those things? But otherwise, this looks good! Thanks! -- 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]
