Noisebridge - User contributions [en]

User:Cortiz

2009-03-12T05:53:11Z

Cortiz:

<h1>C code for the machine learning class 03/11/09</h1>
<p>I used a different scale than the rest of the class. I worked entirely with integers, threshold is 500.</p>
<p>Scroll down for partial output. cheers.</p>

<PRE>
/*=================================================================
Author: Cristian Ortiz (co.ortiz@gmail.com)
Description: Coded for the perceptron class @noisebridge. peace.
===================================================================*/

#include <stdio.h>
#include <stdlib.h>

int dot_prod(int * A, int * B, int n){
int i=0;
int R=0;
for(i=0;i<n;i++){
R+=A[i]*B[i];
}
return R;
}
int * sca_prod(int * A, int n, int x){
int i=0;
int * R;
for(i=0;i<n;i++){
R[i]=A[i]*x;
}
return R;
}
int main(void) {
int input_vec[]={1,0,0,1,0,1,1,1,0,1,1,1};
int expected[]={1,1,1,0};
int input_offset=0;
int w_vec[]={1,1,1};
int learned=0;
int i=0;
int output=0;
int dot_result=0;
int * pvec;
int th=500;
int adjust=0;

while (learned!=4){
//printf("input offset: %d\n", input_offset);
pvec=input_vec+(input_offset*3);
dot_result=dot_prod(pvec, w_vec, 3);
output=(dot_result>th);
printf("ouput: %d supposed to be %d", output, expected[input_offset]);

if(output==expected[input_offset]){printf("\t\t\t<OK!>\n");}
else {printf("\t\t\t<Miss!>\n");}

if(output>expected[input_offset]){adjust=-1;}
else if (output<expected[input_offset]){adjust=1;}
else { learned++; }

if(output!=expected[input_offset]){
for(i=0;i<3;i++){
w_vec[i]=w_vec[i]+(input_vec[i+input_offset*3]*(adjust));
printf("new w_vec[%d]: %d\n", i, w_vec[i]);
}
learned=0;
}
input_offset=(input_offset==3)?0:input_offset+1;
}
for(i=0;i<3;i++){
printf("Final w_vec[%d]=%d\n", i, w_vec[i]);
}
return 0;
}
</PRE>
<h1>Last 20 lines of output:</h1>
<PRE>
ouput: 0 supposed to be 1 <Miss!>
new w_vec[0]: 503
new w_vec[1]: -1
new w_vec[2]: -1
ouput: 1 supposed to be 0 <Miss!>
new w_vec[0]: 502
new w_vec[1]: -2
new w_vec[2]: -2
ouput: 1 supposed to be 1 <OK!>
ouput: 0 supposed to be 1 <Miss!>
new w_vec[0]: 503
new w_vec[1]: -2
new w_vec[2]: -1
ouput: 1 supposed to be 1 <OK!>
ouput: 0 supposed to be 0 <OK!>
ouput: 1 supposed to be 1 <OK!>
ouput: 1 supposed to be 1 <OK!>
Final w_vec[0]=503
Final w_vec[1]=-2
Final w_vec[2]=-1

</PRE>

User:Cortiz

2009-03-12T05:47:00Z

Cortiz:

Didnt know how to make a page so i just pasted it here.
<p>C code for the machine learning class 03/11/09</p>
<p>I used a different scale than the rest of the class. I worked entirely with integers, threshold is 500.</p>
<p>cheers.</p>

<PRE>
/*=================================================================
Author: Cristian Ortiz (co.ortiz@gmail.com)
Description: Coded for the perceptron class @noisebridge. peace.
===================================================================*/

#include <stdio.h>
#include <stdlib.h>

int dot_prod(int * A, int * B, int n){
int i=0;
int R=0;
for(i=0;i<n;i++){
R+=A[i]*B[i];
}
return R;
}
int * sca_prod(int * A, int n, int x){
int i=0;
int * R;
for(i=0;i<n;i++){
R[i]=A[i]*x;
}
return R;
}
int main(void) {
int input_vec[]={1,0,0,1,0,1,1,1,0,1,1,1};
int expected[]={1,1,1,0};
int input_offset=0;
int w_vec[]={1,1,1};
int learned=0;
int i=0;
int output=0;
int dot_result=0;
int * pvec;
int th=500;
int adjust=0;

while (learned!=4){
//printf("input offset: %d\n", input_offset);
pvec=input_vec+(input_offset*3);
dot_result=dot_prod(pvec, w_vec, 3);
output=(dot_result>th);
printf("ouput: %d supposed to be %d", output, expected[input_offset]);

if(output==expected[input_offset]){printf("\t\t\t<OK!>\n");}
else {printf("\t\t\t<Miss!>\n");}

if(output>expected[input_offset]){adjust=-1;}
else if (output<expected[input_offset]){adjust=1;}
else { learned++; }

if(output!=expected[input_offset]){
for(i=0;i<3;i++){
w_vec[i]=w_vec[i]+(input_vec[i+input_offset*3]*(adjust));
printf("new w_vec[%d]: %d\n", i, w_vec[i]);
}
learned=0;
}
input_offset=(input_offset==3)?0:input_offset+1;
}
for(i=0;i<3;i++){
printf("Final w_vec[%d]=%d\n", i, w_vec[i]);
}
return 0;
}
</PRE>

Machine Learning Meetup Notes: 2009-03-11

2009-03-12T05:43:06Z

Cortiz: /* Machine Learning Meetup Notes: 2009-03-11 */

==Machine Learning Meetup Notes: 2009-03-11==

We made perceptrons. To learn how, you should go to Wikipedia [http://en.wikipedia.org/wiki/Perceptron]

We had the following success:

* Mathematica, by Christoph ([https://www.noisebridge.net/wiki/Image:NoisebridgeNeuralNetworks.pdf PDF], because noisebridge is ''a little'' too anti-commercial)
* Matlab/Octave by Jean
* Python by Skory and Rachel
* Ruby by Zhao [https://www.noisebridge.net/wiki/Machine_Learning_Meetup_Notes_Ruby_Zhao]
* C by Cristian [https://www.noisebridge.net/wiki/User:Cortiz]
* [https://www.noisebridge.net/wiki/davids_perceptron.pl Python implementation by David Stainton]

Everyone should upload their code!

User:Cortiz

2009-03-12T05:42:06Z

Cortiz: New page: Didnt know how to make a page so i just pasted it here. code for the machine learning class 03/11/09: <PRE> /*================================================================= Author: ...

Didnt know how to make a page so i just pasted it here.
code for the machine learning class 03/11/09:

<PRE>
/*=================================================================
Author: Cristian Ortiz (co.ortiz@gmail.com)
Description: Coded for the perceptron class @noisebridge. peace.
===================================================================*/

#include <stdio.h>
#include <stdlib.h>

int dot_prod(int * A, int * B, int n){
int i=0;
int R=0;
for(i=0;i<n;i++){
R+=A[i]*B[i];
}
return R;
}
int * sca_prod(int * A, int n, int x){
int i=0;
int * R;
for(i=0;i<n;i++){
R[i]=A[i]*x;
}
return R;
}
int main(void) {
int input_vec[]={1,0,0,1,0,1,1,1,0,1,1,1};
int expected[]={1,1,1,0};
int input_offset=0;
int w_vec[]={1,1,1};
int learned=0;
int i=0;
int output=0;
int dot_result=0;
int * pvec;
int th=500;
int adjust=0;

while (learned!=4){
//printf("input offset: %d\n", input_offset);
pvec=input_vec+(input_offset*3);
dot_result=dot_prod(pvec, w_vec, 3);
output=(dot_result>th);
printf("ouput: %d supposed to be %d", output, expected[input_offset]);

if(output==expected[input_offset]){printf("\t\t\t<OK!>\n");}
else {printf("\t\t\t<Miss!>\n");}

if(output>expected[input_offset]){adjust=-1;}
else if (output<expected[input_offset]){adjust=1;}
else { learned++; }

if(output!=expected[input_offset]){
for(i=0;i<3;i++){
w_vec[i]=w_vec[i]+(input_vec[i+input_offset*3]*(adjust));
printf("new w_vec[%d]: %d\n", i, w_vec[i]);
}
learned=0;
}
input_offset=(input_offset==3)?0:input_offset+1;
}
for(i=0;i<3;i++){
printf("Final w_vec[%d]=%d\n", i, w_vec[i]);
}
return 0;
}
</PRE>

Machine Learning Meetup Notes: 2009-03-11

2009-03-12T05:24:40Z

Cortiz: /* Machine Learning Meetup Notes: 2009-03-11 */

==Machine Learning Meetup Notes: 2009-03-11==

We made perceptrons. To learn how, you should go to Wikipedia [http://en.wikipedia.org/wiki/Perceptron]

We had the following success:

* Mathematica, by Christoph ([https://www.noisebridge.net/wiki/Image:NoisebridgeNeuralNetworks.pdf PDF], because noisebridge is ''a little'' too anti-commercial)
* Matlab/Octave by Jean
* Python by Skory and Rachel
* Ruby by Zhao [https://www.noisebridge.net/wiki/Machine_Learning_Meetup_Notes_Ruby_Zhao]
* C by Cristian
* [https://www.noisebridge.net/wiki/davids_perceptron.pl Python implementation by David Stainton]

Everyone should upload their code!