You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@madlib.apache.org by "Feng, Xixuan (Aaron)" <aa...@recruit.ai> on 2017/02/22 02:33:02 UTC
Re: Regarding new Perceptron module
Hi Auon,
I have made some effort to try implementing multi-layer perceptron a few
years back but never got the time to finish it. Latest commits on these 2
branches has the history
https://github.com/haying/madlib/commits/nn_design
https://github.com/haying/madlib/commits/mlp
I were using the convex programming framework in MADlib to implement it
with stochastic gradient descent. You may or may not do it in the same way.
Thanks,
Feng, Xixuan (Aaron)
On Wed, Feb 22, 2017 at 8:59 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
> Hi all,
>
> I am currently working on implementing Perceptron Learning Algorithm in
> Madlib. This module would serve the purpose of a single neuron when
> implementing any multi-layer neural network.
>
> Since, it involve state transitions, optimizations,etc., could you guys
> suggest a module that could help a little in this context?
>
>
>
>
>
>
> Regards,
>
> Auon
>
--
- Aaron
Re: Regarding new Perceptron module
Posted by "Kazmi,Auon H" <ak...@ufl.edu>.
Thanks Aaron!
Auon
________________________________
From: Feng, Xixuan (Aaron) <aa...@recruit.ai>
Sent: Thursday, March 2, 2017 6:18:53 PM
To: dev@madlib.incubator.apache.org
Subject: Re: Regarding new Perceptron module
Probably. You could refer to the SGD version of other modules, like SVM or
Logistic Regression.
- Aaron
2017年3月3日(金) 8:14 Kazmi,Auon H <ak...@ufl.edu>:
> Hey Xixuan,
>
> I couldn't find your python and postgres files for MLP. Did you just work
> on C++ layer?
>
>
>
> Auon
>
> ________________________________
> From: Rahul Iyer <ra...@gmail.com>
> Sent: Wednesday, February 22, 2017 4:12:18 PM
> To: dev@madlib.incubator.apache.org
> Subject: Re: Regarding new Perceptron module
>
> I believe building a backprop-learned MLP would be much better than using
> the perceptron learning algo. In my experience, perceptron learning works
> quite slow; happy to hear other thoughts.
>
> On Wed, Feb 22, 2017 at 11:40 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
>
> > Hi Xixuan,
> >
> > THanks! Is starting with single layer perceptron (and thus using
> > perceptron learning algo rather than SGD) a better idea?
> >
> >
> >
> > Auon
> >
> > ________________________________
> > From: Feng, Xixuan (Aaron) <aa...@recruit.ai>
> > Sent: Tuesday, February 21, 2017 9:33:02 PM
> > To: dev@madlib.incubator.apache.org
> > Subject: Re: Regarding new Perceptron module
> >
> > Hi Auon,
> >
> > I have made some effort to try implementing multi-layer perceptron a few
> > years back but never got the time to finish it. Latest commits on these 2
> > branches has the history
> > https://github.com/haying/madlib/commits/nn_design
> > https://github.com/haying/madlib/commits/mlp
> >
> > I were using the convex programming framework in MADlib to implement it
> > with stochastic gradient descent. You may or may not do it in the same
> way.
> >
> > Thanks,
> > Feng, Xixuan (Aaron)
> >
> > On Wed, Feb 22, 2017 at 8:59 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
> >
> > > Hi all,
> > >
> > > I am currently working on implementing Perceptron Learning Algorithm in
> > > Madlib. This module would serve the purpose of a single neuron when
> > > implementing any multi-layer neural network.
> > >
> > > Since, it involve state transitions, optimizations,etc., could you
> guys
> > > suggest a module that could help a little in this context?
> > >
> > >
> > >
> > >
> > >
> > >
> > > Regards,
> > >
> > > Auon
> > >
> >
> >
> >
> > --
> > - Aaron
> >
>
--
- Aaron
Re: Regarding new Perceptron module
Posted by "Feng, Xixuan (Aaron)" <aa...@recruit.ai>.
Probably. You could refer to the SGD version of other modules, like SVM or
Logistic Regression.
- Aaron
2017年3月3日(金) 8:14 Kazmi,Auon H <ak...@ufl.edu>:
> Hey Xixuan,
>
> I couldn't find your python and postgres files for MLP. Did you just work
> on C++ layer?
>
>
>
> Auon
>
> ________________________________
> From: Rahul Iyer <ra...@gmail.com>
> Sent: Wednesday, February 22, 2017 4:12:18 PM
> To: dev@madlib.incubator.apache.org
> Subject: Re: Regarding new Perceptron module
>
> I believe building a backprop-learned MLP would be much better than using
> the perceptron learning algo. In my experience, perceptron learning works
> quite slow; happy to hear other thoughts.
>
> On Wed, Feb 22, 2017 at 11:40 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
>
> > Hi Xixuan,
> >
> > THanks! Is starting with single layer perceptron (and thus using
> > perceptron learning algo rather than SGD) a better idea?
> >
> >
> >
> > Auon
> >
> > ________________________________
> > From: Feng, Xixuan (Aaron) <aa...@recruit.ai>
> > Sent: Tuesday, February 21, 2017 9:33:02 PM
> > To: dev@madlib.incubator.apache.org
> > Subject: Re: Regarding new Perceptron module
> >
> > Hi Auon,
> >
> > I have made some effort to try implementing multi-layer perceptron a few
> > years back but never got the time to finish it. Latest commits on these 2
> > branches has the history
> > https://github.com/haying/madlib/commits/nn_design
> > https://github.com/haying/madlib/commits/mlp
> >
> > I were using the convex programming framework in MADlib to implement it
> > with stochastic gradient descent. You may or may not do it in the same
> way.
> >
> > Thanks,
> > Feng, Xixuan (Aaron)
> >
> > On Wed, Feb 22, 2017 at 8:59 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
> >
> > > Hi all,
> > >
> > > I am currently working on implementing Perceptron Learning Algorithm in
> > > Madlib. This module would serve the purpose of a single neuron when
> > > implementing any multi-layer neural network.
> > >
> > > Since, it involve state transitions, optimizations,etc., could you
> guys
> > > suggest a module that could help a little in this context?
> > >
> > >
> > >
> > >
> > >
> > >
> > > Regards,
> > >
> > > Auon
> > >
> >
> >
> >
> > --
> > - Aaron
> >
>
--
- Aaron
Re: Regarding new Perceptron module
Posted by "Kazmi,Auon H" <ak...@ufl.edu>.
Hey Xixuan,
I couldn't find your python and postgres files for MLP. Did you just work on C++ layer?
Auon
________________________________
From: Rahul Iyer <ra...@gmail.com>
Sent: Wednesday, February 22, 2017 4:12:18 PM
To: dev@madlib.incubator.apache.org
Subject: Re: Regarding new Perceptron module
I believe building a backprop-learned MLP would be much better than using
the perceptron learning algo. In my experience, perceptron learning works
quite slow; happy to hear other thoughts.
On Wed, Feb 22, 2017 at 11:40 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
> Hi Xixuan,
>
> THanks! Is starting with single layer perceptron (and thus using
> perceptron learning algo rather than SGD) a better idea?
>
>
>
> Auon
>
> ________________________________
> From: Feng, Xixuan (Aaron) <aa...@recruit.ai>
> Sent: Tuesday, February 21, 2017 9:33:02 PM
> To: dev@madlib.incubator.apache.org
> Subject: Re: Regarding new Perceptron module
>
> Hi Auon,
>
> I have made some effort to try implementing multi-layer perceptron a few
> years back but never got the time to finish it. Latest commits on these 2
> branches has the history
> https://github.com/haying/madlib/commits/nn_design
> https://github.com/haying/madlib/commits/mlp
>
> I were using the convex programming framework in MADlib to implement it
> with stochastic gradient descent. You may or may not do it in the same way.
>
> Thanks,
> Feng, Xixuan (Aaron)
>
> On Wed, Feb 22, 2017 at 8:59 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
>
> > Hi all,
> >
> > I am currently working on implementing Perceptron Learning Algorithm in
> > Madlib. This module would serve the purpose of a single neuron when
> > implementing any multi-layer neural network.
> >
> > Since, it involve state transitions, optimizations,etc., could you guys
> > suggest a module that could help a little in this context?
> >
> >
> >
> >
> >
> >
> > Regards,
> >
> > Auon
> >
>
>
>
> --
> - Aaron
>
Re: Regarding new Perceptron module
Posted by Rahul Iyer <ra...@gmail.com>.
I believe building a backprop-learned MLP would be much better than using
the perceptron learning algo. In my experience, perceptron learning works
quite slow; happy to hear other thoughts.
On Wed, Feb 22, 2017 at 11:40 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
> Hi Xixuan,
>
> THanks! Is starting with single layer perceptron (and thus using
> perceptron learning algo rather than SGD) a better idea?
>
>
>
> Auon
>
> ________________________________
> From: Feng, Xixuan (Aaron) <aa...@recruit.ai>
> Sent: Tuesday, February 21, 2017 9:33:02 PM
> To: dev@madlib.incubator.apache.org
> Subject: Re: Regarding new Perceptron module
>
> Hi Auon,
>
> I have made some effort to try implementing multi-layer perceptron a few
> years back but never got the time to finish it. Latest commits on these 2
> branches has the history
> https://github.com/haying/madlib/commits/nn_design
> https://github.com/haying/madlib/commits/mlp
>
> I were using the convex programming framework in MADlib to implement it
> with stochastic gradient descent. You may or may not do it in the same way.
>
> Thanks,
> Feng, Xixuan (Aaron)
>
> On Wed, Feb 22, 2017 at 8:59 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
>
> > Hi all,
> >
> > I am currently working on implementing Perceptron Learning Algorithm in
> > Madlib. This module would serve the purpose of a single neuron when
> > implementing any multi-layer neural network.
> >
> > Since, it involve state transitions, optimizations,etc., could you guys
> > suggest a module that could help a little in this context?
> >
> >
> >
> >
> >
> >
> > Regards,
> >
> > Auon
> >
>
>
>
> --
> - Aaron
>
Re: Regarding new Perceptron module
Posted by "Kazmi,Auon H" <ak...@ufl.edu>.
Hi Xixuan,
THanks! Is starting with single layer perceptron (and thus using perceptron learning algo rather than SGD) a better idea?
Auon
________________________________
From: Feng, Xixuan (Aaron) <aa...@recruit.ai>
Sent: Tuesday, February 21, 2017 9:33:02 PM
To: dev@madlib.incubator.apache.org
Subject: Re: Regarding new Perceptron module
Hi Auon,
I have made some effort to try implementing multi-layer perceptron a few
years back but never got the time to finish it. Latest commits on these 2
branches has the history
https://github.com/haying/madlib/commits/nn_design
https://github.com/haying/madlib/commits/mlp
I were using the convex programming framework in MADlib to implement it
with stochastic gradient descent. You may or may not do it in the same way.
Thanks,
Feng, Xixuan (Aaron)
On Wed, Feb 22, 2017 at 8:59 AM, Kazmi,Auon H <ak...@ufl.edu> wrote:
> Hi all,
>
> I am currently working on implementing Perceptron Learning Algorithm in
> Madlib. This module would serve the purpose of a single neuron when
> implementing any multi-layer neural network.
>
> Since, it involve state transitions, optimizations,etc., could you guys
> suggest a module that could help a little in this context?
>
>
>
>
>
>
> Regards,
>
> Auon
>
--
- Aaron