Car-Recognition icon indicating copy to clipboard operation
Car-Recognition copied to clipboard

model.prediect() is too slow when runing demo.py

Open Shame-fight opened this issue 3 years ago • 9 comments

It can be used 8 to 10 minutes just in this step(model.predict),I do not know the reason ,who can help me. my environment is anaconda+tensorflow1.14+keras2.2.5+CUDN10+cuDNN7.4+windows10,no visual studio(is the reason?) Hope to hear any reply.

Shame-fight avatar Apr 20 '21 03:04 Shame-fight

How many images are you using in demo?

Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10

From: @.> Sent: Tuesday, April 20, 2021 8:03 AM To: @.> Cc: @.***> Subject: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27)

It can be used 8 to 10 minutes just in this step(model.predict),I do not know the reason ,who can help me. my environment is anaconda+tensorflow1.14+keras2.2.5+CUDN10+cuDNN7.4+windows10,no visual studio(is the reason?) Hope to hear any reply.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/foamliu/Car-Recognition/issues/27, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MUNC6MP5XHSRRTFTOTTJTVG3ANCNFSM43HHZEIA.

Usman-Ghani123 avatar Apr 20 '21 04:04 Usman-Ghani123

@Usman-Ghani123 just one image,and my windows has intel i7+rtx3090

Shame-fight avatar Apr 20 '21 05:04 Shame-fight

Is it giving correct result?

Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10

From: @.> Sent: Tuesday, April 20, 2021 10:21 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27)

@Usman-Ghani123https://github.com/Usman-Ghani123 just one image,and my windows has intel i7+rtx3090

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/foamliu/Car-Recognition/issues/27#issuecomment-822981298, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MQHCWZUZNEVYCYE6PTTJUFN7ANCNFSM43HHZEIA.

Usman-Ghani123 avatar Apr 20 '21 05:04 Usman-Ghani123

@Usman-Ghani123 yes ,but the probility is "nan".

Shame-fight avatar Apr 20 '21 05:04 Shame-fight

I guess there is some problem in probability which is causing your demo to be slow.

Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10

From: @.> Sent: Tuesday, April 20, 2021 10:45 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27)

@Usman-Ghani123https://github.com/Usman-Ghani123 yes ,but the probility is "nan".

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/foamliu/Car-Recognition/issues/27#issuecomment-822990426, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MXOICXYIR3GAMJNSG3TJUIIHANCNFSM43HHZEIA.

Usman-Ghani123 avatar Apr 20 '21 05:04 Usman-Ghani123

I guess there is some problem in probability which is causing your demo to be slow. Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10 From: @.> Sent: Tuesday, April 20, 2021 10:45 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27) @Usman-Ghani123https://github.com/Usman-Ghani123 yes ,but the probility is "nan". — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#27 (comment)>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MXOICXYIR3GAMJNSG3TJUIIHANCNFSM43HHZEIA.

yes,what should I do can help solving this problem?

Shame-fight avatar Apr 20 '21 05:04 Shame-fight

Are you running the same code as in the GitHub file?

Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10

From: @.> Sent: Tuesday, April 20, 2021 10:57 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27)

I guess there is some problem in probability which is causing your demo to be slow. Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10 From: @.> Sent: Tuesday, April 20, 2021 10:45 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27https://github.com/foamliu/Car-Recognition/issues/27) @Usman-Ghani123https://github.com/Usman-Ghani123https://github.com/Usman-Ghani123 yes ,but the probility is "nan". — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#27 (comment)https://github.com/foamliu/Car-Recognition/issues/27#issuecomment-822990426>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MXOICXYIR3GAMJNSG3TJUIIHANCNFSM43HHZEIA.

yes,what should I do can help solving this problem?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/foamliu/Car-Recognition/issues/27#issuecomment-822995112, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MQT3GD2Q44POREUTVDTJUJTJANCNFSM43HHZEIA.

Usman-Ghani123 avatar Apr 20 '21 05:04 Usman-Ghani123

Are you running the same code as in the GitHub file? Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10 From: @.> Sent: Tuesday, April 20, 2021 10:57 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27) I guess there is some problem in probability which is causing your demo to be slow. Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10 From: @.> Sent: Tuesday, April 20, 2021 10:45 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27<#27>) @Usman-Ghani123https://github.com/Usman-Ghani123https://github.com/Usman-Ghani123 yes ,but the probility is "nan". — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#27 (comment)<#27 (comment)>>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MXOICXYIR3GAMJNSG3TJUIIHANCNFSM43HHZEIA. yes,what should I do can help solving this problem? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#27 (comment)>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MQT3GD2Q44POREUTVDTJUJTJANCNFSM43HHZEIA.

I only modified the code about loading images in demo.py, because I only need to test one image. At the same time, I did not download the data set. Sorry, I cannot provide the code for some reasons. I apologize for the inconvenience caused. 11

Shame-fight avatar Apr 20 '21 06:04 Shame-fight

Print the array of pred and print the class_id and print prob as well

It will give a clear idea that are we getting any value or not Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10

From: @.> Sent: Tuesday, April 20, 2021 11:20 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27)

Are you running the same code as in the GitHub file? Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10 From: @.> Sent: Tuesday, April 20, 2021 10:57 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27https://github.com/foamliu/Car-Recognition/issues/27) I guess there is some problem in probability which is causing your demo to be slow. Sent from Mailhttps://go.microsoft.com/fwlink/?LinkId=550986 for Windows 10 From: @.> Sent: Tuesday, April 20, 2021 10:45 AM To: @.> Cc: Syed Muhmmad Usman @.>; @.> Subject: Re: [foamliu/Car-Recognition] model.prediect() is too slow when runing demo.py (#27https://github.com/foamliu/Car-Recognition/issues/27<#27https://github.com/foamliu/Car-Recognition/issues/27>) @Usman-Ghani123https://github.com/Usman-Ghani123https://github.com/Usman-Ghani123https://github.com/Usman-Ghani123 yes ,but the probility is "nan". — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#27https://github.com/foamliu/Car-Recognition/issues/27 (comment)<#27 (comment)https://github.com/foamliu/Car-Recognition/issues/27#issuecomment-822990426>>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MXOICXYIR3GAMJNSG3TJUIIHANCNFSM43HHZEIA. yes,what should I do can help solving this problem? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#27 (comment)https://github.com/foamliu/Car-Recognition/issues/27#issuecomment-822995112>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MQT3GD2Q44POREUTVDTJUJTJANCNFSM43HHZEIA.

I only modified the code about loading images in demo.py, because I only need to test one image. At the same time, I did not download the data set. Sorry, I cannot provide the code for some reasons. I apologize for the inconvenience caused. [11]https://user-images.githubusercontent.com/30313218/115346859-4d368980-a1e3-11eb-945b-0f75f8695774.jpg

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/foamliu/Car-Recognition/issues/27#issuecomment-823005896, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AO2V5MTC4IXSCX3CENH55GTTJUMLVANCNFSM43HHZEIA.

Usman-Ghani123 avatar Apr 20 '21 06:04 Usman-Ghani123