Reducing time of deep learning method

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Reducing time of deep learning method

Edward Wiskers
Hi all, 

Kindly I need to preprocess and build a deep learning model to classify my data, but operations execution are taking so long time. Could you advise on how to reduce the learning process time, in general, of deep learning method?

Thank you. 

Edward 



_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|

Re: Reducing time of deep learning method

Eibe Frank-2
Administrator
Do you have a discrete GPU in your computer? That is the standard way to accelerate DL if you want to run it on your own machine.

In recent versions of WekaDeeplearning4j, you need to run a command-line script to enable GPU support:

  https://deeplearning.cms.waikato.ac.nz/install/

Note that most libraries, including WekaDeeplearning4j, require an NVIDIA GPU. For WekaDeeplearning4j, you need to separately install NVIDIA’s CUDA library on your computer (see above link).

Cheers,
Eibe

> On 13/09/2019, at 12:13 PM, Edward Wiskers <[hidden email]> wrote:
>
> Hi all,
>
> Kindly I need to preprocess and build a deep learning model to classify my data, but operations execution are taking so long time. Could you advise on how to reduce the learning process time, in general, of deep learning method?
>
> Thank you.
>
> Edward
>
>
> _______________________________________________
> Wekalist mailing list -- [hidden email]
> Send posts to: To unsubscribe send an email to [hidden email]
> To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|

Re: Reducing time of deep learning method

Edward Wiskers
Hi Eibe, 

Thank you for the prompt reply. 

I am not having the discrete GPU, but do you believe that sampling (without replacement) can be useful to be applied before passing the data into DL? Is there any preprocess method that may reduce the speed of DL algorithm?

Best, 
Edward

On Fri, Sep 13, 2019 at 8:19 AM Eibe Frank <[hidden email]> wrote:
Do you have a discrete GPU in your computer? That is the standard way to accelerate DL if you want to run it on your own machine.

In recent versions of WekaDeeplearning4j, you need to run a command-line script to enable GPU support:

  https://deeplearning.cms.waikato.ac.nz/install/

Note that most libraries, including WekaDeeplearning4j, require an NVIDIA GPU. For WekaDeeplearning4j, you need to separately install NVIDIA’s CUDA library on your computer (see above link).

Cheers,
Eibe

> On 13/09/2019, at 12:13 PM, Edward Wiskers <[hidden email]> wrote:
>
> Hi all,
>
> Kindly I need to preprocess and build a deep learning model to classify my data, but operations execution are taking so long time. Could you advise on how to reduce the learning process time, in general, of deep learning method?
>
> Thank you.
>
> Edward
>
>
> _______________________________________________
> Wekalist mailing list -- [hidden email]
> Send posts to: To unsubscribe send an email to [hidden email]
> To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|

Re: Reducing time of deep learning method

Edward Wiskers
Hi Eibe, 

Thank you for the prompt reply. 

I am not having the discrete GPU, but do you believe that sampling (without replacement) can be useful to be applied before passing the data into DL? Is there any preprocess method that may reduce the speed of DL algorithm?

Best, 
Edward

On Fri, 13 Sep 2019, 9:31 am Edward Wiskers, <[hidden email]> wrote:
Hi Eibe, 

Thank you for the prompt reply. 

I am not having the discrete GPU, but do you believe that sampling (without replacement) can be useful to be applied before passing the data into DL? Is there any preprocess method that may reduce the speed of DL algorithm?

Best, 
Edward

On Fri, Sep 13, 2019 at 8:19 AM Eibe Frank <[hidden email]> wrote:
Do you have a discrete GPU in your computer? That is the standard way to accelerate DL if you want to run it on your own machine.

In recent versions of WekaDeeplearning4j, you need to run a command-line script to enable GPU support:

  https://deeplearning.cms.waikato.ac.nz/install/

Note that most libraries, including WekaDeeplearning4j, require an NVIDIA GPU. For WekaDeeplearning4j, you need to separately install NVIDIA’s CUDA library on your computer (see above link).

Cheers,
Eibe

> On 13/09/2019, at 12:13 PM, Edward Wiskers <[hidden email]> wrote:
>
> Hi all,
>
> Kindly I need to preprocess and build a deep learning model to classify my data, but operations execution are taking so long time. Could you advise on how to reduce the learning process time, in general, of deep learning method?
>
> Thank you.
>
> Edward
>
>
> _______________________________________________
> Wekalist mailing list -- [hidden email]
> Send posts to: To unsubscribe send an email to [hidden email]
> To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|

Re: Reducing time of deep learning method

Eibe Frank-2
Administrator
What kind of network are you using? What is the size of the mini-batches you are using for mini-batch stochastic gradient descent?

Cheers,
Eibe

On Sat, Sep 14, 2019 at 2:52 PM Edward Wiskers <[hidden email]> wrote:
Hi Eibe, 

Thank you for the prompt reply. 

I am not having the discrete GPU, but do you believe that sampling (without replacement) can be useful to be applied before passing the data into DL? Is there any preprocess method that may reduce the speed of DL algorithm?

Best, 
Edward

On Fri, 13 Sep 2019, 9:31 am Edward Wiskers, <[hidden email]> wrote:
Hi Eibe, 

Thank you for the prompt reply. 

I am not having the discrete GPU, but do you believe that sampling (without replacement) can be useful to be applied before passing the data into DL? Is there any preprocess method that may reduce the speed of DL algorithm?

Best, 
Edward

On Fri, Sep 13, 2019 at 8:19 AM Eibe Frank <[hidden email]> wrote:
Do you have a discrete GPU in your computer? That is the standard way to accelerate DL if you want to run it on your own machine.

In recent versions of WekaDeeplearning4j, you need to run a command-line script to enable GPU support:

  https://deeplearning.cms.waikato.ac.nz/install/

Note that most libraries, including WekaDeeplearning4j, require an NVIDIA GPU. For WekaDeeplearning4j, you need to separately install NVIDIA’s CUDA library on your computer (see above link).

Cheers,
Eibe

> On 13/09/2019, at 12:13 PM, Edward Wiskers <[hidden email]> wrote:
>
> Hi all,
>
> Kindly I need to preprocess and build a deep learning model to classify my data, but operations execution are taking so long time. Could you advise on how to reduce the learning process time, in general, of deep learning method?
>
> Thank you.
>
> Edward
>
>
> _______________________________________________
> Wekalist mailing list -- [hidden email]
> Send posts to: To unsubscribe send an email to [hidden email]
> To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list -- [hidden email]
Send posts to: To unsubscribe send an email to [hidden email]
To subscribe, unsubscribe, etc., visit %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html