Mixing
How to compute distance between two vectors based on measures?
I want to compute the average of distances between each column and other columns in a matrix except itself.
I wrote this code:
For i=1:m
For j=1:m-1
If(i==j)
d=0;
Else
d=pdist2(a(:,i),a(:,j),'jaccard');
s=sum(d)/(m-1);
End
End
End
The matrix is nxm.
I know that pdist2 function support some distance measures like jaccard , cosine , hamming , euclidean....
But If I want to compute canberra or clark distance measures or measures like them, I should define a function in pdist2 this way:
D=pdist2(x,y,@distfun)
I wanted to know how to define a distance function for these measures?
Edit: The aim is to then empirically obtain the point where the values start to decrease but the figures don't show this...
matlab matrix euclidean-distance
edited Jan 4 at 9:50
Wolfie
17k51746
asked Jan 2 at 21:59
engineer100engineer100
105
add a comment |
I want to compute the average of distances between each column and other columns in a matrix except itself.
I wrote this code:
For i=1:m
For j=1:m-1
If(i==j)
d=0;
Else
d=pdist2(a(:,i),a(:,j),'jaccard');
s=sum(d)/(m-1);
End
End
End
The matrix is nxm.
I know that pdist2 function support some distance measures like jaccard , cosine , hamming , euclidean....
But If I want to compute canberra or clark distance measures or measures like them, I should define a function in pdist2 this way:
D=pdist2(x,y,@distfun)
I wanted to know how to define a distance function for these measures?
Edit: The aim is to then empirically obtain the point where the values start to decrease but the figures don't show this...
matlab matrix euclidean-distance
edited Jan 4 at 9:50
Wolfie
17k51746
asked Jan 2 at 21:59
engineer100engineer100
105
add a comment |
I want to compute the average of distances between each column and other columns in a matrix except itself.
I wrote this code:
For i=1:m
For j=1:m-1
If(i==j)
d=0;
Else
d=pdist2(a(:,i),a(:,j),'jaccard');
s=sum(d)/(m-1);
End
End
End
The matrix is nxm.
I know that pdist2 function support some distance measures like jaccard , cosine , hamming , euclidean....
But If I want to compute canberra or clark distance measures or measures like them, I should define a function in pdist2 this way:
D=pdist2(x,y,@distfun)
I wanted to know how to define a distance function for these measures?
Edit: The aim is to then empirically obtain the point where the values start to decrease but the figures don't show this...
matlab matrix euclidean-distance
edited Jan 4 at 9:50
Wolfie
17k51746
asked Jan 2 at 21:59
engineer100engineer100
105
I want to compute the average of distances between each column and other columns in a matrix except itself.
I wrote this code:
For i=1:m
For j=1:m-1
If(i==j)
d=0;
Else
d=pdist2(a(:,i),a(:,j),'jaccard');
s=sum(d)/(m-1);
End
End
End
The matrix is nxm.
I know that pdist2 function support some distance measures like jaccard , cosine , hamming , euclidean....
But If I want to compute canberra or clark distance measures or measures like them, I should define a function in pdist2 this way:
D=pdist2(x,y,@distfun)
I wanted to know how to define a distance function for these measures?
Edit: The aim is to then empirically obtain the point where the values start to decrease but the figures don't show this...
matlab matrix euclidean-distance
matlab matrix euclidean-distance
edited Jan 4 at 9:50
Wolfie
17k51746
asked Jan 2 at 21:59
engineer100engineer100
105
edited Jan 4 at 9:50
Wolfie
17k51746
asked Jan 2 at 21:59
engineer100engineer100
105
edited Jan 4 at 9:50
Wolfie
17k51746
edited Jan 4 at 9:50
Wolfie
17k51746
edited Jan 4 at 9:50
Wolfie
17k51746
17k51746
asked Jan 2 at 21:59
engineer100engineer100
105
asked Jan 2 at 21:59
engineer100engineer100
105
asked Jan 2 at 21:59
engineer100engineer100
105
105
add a comment |
add a comment |
1 Answer
1
active
oldest
votes
The Canberra Distance is defined as
You are passing two column vectors into pdist2, so we can implement the above equation as an anonymous function which accepts to vector inputs...
fCanberraDist = @(p,q) sum( abs(p - q) ./ ( abs(p) + abs(q) ) );
Now you have the function handle, you can use the following:
% ... previous code
d = pdist2( a(:,i), a(:,j), fCanberraDist );
% ... further code
Note you're not currently doing anything with your result, I'm assuming that's just for the sake of the example code.
You can define any other custom distance measure in a similar way, as either an anonymous function or a fully separate function (in its own .m file or defined locally).
answered Jan 3 at 9:29
WolfieWolfie
17k51746
After the loop, I want to use plot(s) to plot the m disrances and maybe I can find a threshold for these values.Is It practical to find a threshold from the figure of the values?or there is a better procedure?
– engineer100
Jan 4 at 8:00
It's unclear what you mean by "find a threshold" - what would be the criteria for this threshold? If you can define it mathematically then it's better to do programmatically.
– Wolfie
Jan 4 at 8:07
I want to empirically obtain the point where the values start to decrease but the figures don't show this, for example, I attached the figure of Canberra distances in my question. You mean that I should define It mathematically first.
– engineer100
Jan 4 at 9:00
I have answered your original question, I can't tell you why a figure you've produced doesn't show the results you expected... You've not given a Minimal, Complete, and Verifiable example to reproduce these results, or any explanation as to why they aren't as expected? This perhaps needs its own question, if the original question here about implementing a custom distance measure has been answered. Please mark my answer as accepted if this is the case.
– Wolfie
Jan 4 at 10:21
add a comment |
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f54013724%2fhow-to-compute-distance-between-two-vectors-based-on-measures%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
The Canberra Distance is defined as
You are passing two column vectors into pdist2, so we can implement the above equation as an anonymous function which accepts to vector inputs...
fCanberraDist = @(p,q) sum( abs(p - q) ./ ( abs(p) + abs(q) ) );
Now you have the function handle, you can use the following:
% ... previous code
d = pdist2( a(:,i), a(:,j), fCanberraDist );
% ... further code
Note you're not currently doing anything with your result, I'm assuming that's just for the sake of the example code.
You can define any other custom distance measure in a similar way, as either an anonymous function or a fully separate function (in its own .m file or defined locally).
answered Jan 3 at 9:29
WolfieWolfie
17k51746
After the loop, I want to use plot(s) to plot the m disrances and maybe I can find a threshold for these values.Is It practical to find a threshold from the figure of the values?or there is a better procedure?
– engineer100
Jan 4 at 8:00
It's unclear what you mean by "find a threshold" - what would be the criteria for this threshold? If you can define it mathematically then it's better to do programmatically.
– Wolfie
Jan 4 at 8:07
I want to empirically obtain the point where the values start to decrease but the figures don't show this, for example, I attached the figure of Canberra distances in my question. You mean that I should define It mathematically first.
– engineer100
Jan 4 at 9:00
I have answered your original question, I can't tell you why a figure you've produced doesn't show the results you expected... You've not given a Minimal, Complete, and Verifiable example to reproduce these results, or any explanation as to why they aren't as expected? This perhaps needs its own question, if the original question here about implementing a custom distance measure has been answered. Please mark my answer as accepted if this is the case.
– Wolfie
Jan 4 at 10:21
add a comment |
The Canberra Distance is defined as
You are passing two column vectors into pdist2, so we can implement the above equation as an anonymous function which accepts to vector inputs...
fCanberraDist = @(p,q) sum( abs(p - q) ./ ( abs(p) + abs(q) ) );
Now you have the function handle, you can use the following:
% ... previous code
d = pdist2( a(:,i), a(:,j), fCanberraDist );
% ... further code
Note you're not currently doing anything with your result, I'm assuming that's just for the sake of the example code.
You can define any other custom distance measure in a similar way, as either an anonymous function or a fully separate function (in its own .m file or defined locally).
answered Jan 3 at 9:29
WolfieWolfie
17k51746
After the loop, I want to use plot(s) to plot the m disrances and maybe I can find a threshold for these values.Is It practical to find a threshold from the figure of the values?or there is a better procedure?
– engineer100
Jan 4 at 8:00
It's unclear what you mean by "find a threshold" - what would be the criteria for this threshold? If you can define it mathematically then it's better to do programmatically.
– Wolfie
Jan 4 at 8:07
I want to empirically obtain the point where the values start to decrease but the figures don't show this, for example, I attached the figure of Canberra distances in my question. You mean that I should define It mathematically first.
– engineer100
Jan 4 at 9:00
I have answered your original question, I can't tell you why a figure you've produced doesn't show the results you expected... You've not given a Minimal, Complete, and Verifiable example to reproduce these results, or any explanation as to why they aren't as expected? This perhaps needs its own question, if the original question here about implementing a custom distance measure has been answered. Please mark my answer as accepted if this is the case.
– Wolfie
Jan 4 at 10:21
add a comment |
The Canberra Distance is defined as
You are passing two column vectors into pdist2, so we can implement the above equation as an anonymous function which accepts to vector inputs...
fCanberraDist = @(p,q) sum( abs(p - q) ./ ( abs(p) + abs(q) ) );
Now you have the function handle, you can use the following:
% ... previous code
d = pdist2( a(:,i), a(:,j), fCanberraDist );
% ... further code
Note you're not currently doing anything with your result, I'm assuming that's just for the sake of the example code.
You can define any other custom distance measure in a similar way, as either an anonymous function or a fully separate function (in its own .m file or defined locally).
answered Jan 3 at 9:29
WolfieWolfie
17k51746
The Canberra Distance is defined as
You are passing two column vectors into pdist2, so we can implement the above equation as an anonymous function which accepts to vector inputs...
fCanberraDist = @(p,q) sum( abs(p - q) ./ ( abs(p) + abs(q) ) );
Now you have the function handle, you can use the following:
% ... previous code
d = pdist2( a(:,i), a(:,j), fCanberraDist );
% ... further code
Note you're not currently doing anything with your result, I'm assuming that's just for the sake of the example code.
You can define any other custom distance measure in a similar way, as either an anonymous function or a fully separate function (in its own .m file or defined locally).
answered Jan 3 at 9:29
WolfieWolfie
17k51746
answered Jan 3 at 9:29
WolfieWolfie
17k51746
answered Jan 3 at 9:29
WolfieWolfie
17k51746
answered Jan 3 at 9:29
WolfieWolfie
17k51746
17k51746
After the loop, I want to use plot(s) to plot the m disrances and maybe I can find a threshold for these values.Is It practical to find a threshold from the figure of the values?or there is a better procedure?
– engineer100
Jan 4 at 8:00
It's unclear what you mean by "find a threshold" - what would be the criteria for this threshold? If you can define it mathematically then it's better to do programmatically.
– Wolfie
Jan 4 at 8:07
I want to empirically obtain the point where the values start to decrease but the figures don't show this, for example, I attached the figure of Canberra distances in my question. You mean that I should define It mathematically first.
– engineer100
Jan 4 at 9:00
I have answered your original question, I can't tell you why a figure you've produced doesn't show the results you expected... You've not given a Minimal, Complete, and Verifiable example to reproduce these results, or any explanation as to why they aren't as expected? This perhaps needs its own question, if the original question here about implementing a custom distance measure has been answered. Please mark my answer as accepted if this is the case.
– Wolfie
Jan 4 at 10:21
add a comment |
After the loop, I want to use plot(s) to plot the m disrances and maybe I can find a threshold for these values.Is It practical to find a threshold from the figure of the values?or there is a better procedure?
– engineer100
Jan 4 at 8:00
It's unclear what you mean by "find a threshold" - what would be the criteria for this threshold? If you can define it mathematically then it's better to do programmatically.
– Wolfie
Jan 4 at 8:07
I want to empirically obtain the point where the values start to decrease but the figures don't show this, for example, I attached the figure of Canberra distances in my question. You mean that I should define It mathematically first.
– engineer100
Jan 4 at 9:00
I have answered your original question, I can't tell you why a figure you've produced doesn't show the results you expected... You've not given a Minimal, Complete, and Verifiable example to reproduce these results, or any explanation as to why they aren't as expected? This perhaps needs its own question, if the original question here about implementing a custom distance measure has been answered. Please mark my answer as accepted if this is the case.
– Wolfie
Jan 4 at 10:21
After the loop, I want to use plot(s) to plot the m disrances and maybe I can find a threshold for these values.Is It practical to find a threshold from the figure of the values?or there is a better procedure?
– engineer100
Jan 4 at 8:00
After the loop, I want to use plot(s) to plot the m disrances and maybe I can find a threshold for these values.Is It practical to find a threshold from the figure of the values?or there is a better procedure?
– engineer100
Jan 4 at 8:00
It's unclear what you mean by "find a threshold" - what would be the criteria for this threshold? If you can define it mathematically then it's better to do programmatically.
– Wolfie
Jan 4 at 8:07
It's unclear what you mean by "find a threshold" - what would be the criteria for this threshold? If you can define it mathematically then it's better to do programmatically.
– Wolfie
Jan 4 at 8:07
I want to empirically obtain the point where the values start to decrease but the figures don't show this, for example, I attached the figure of Canberra distances in my question. You mean that I should define It mathematically first.
– engineer100
Jan 4 at 9:00
I want to empirically obtain the point where the values start to decrease but the figures don't show this, for example, I attached the figure of Canberra distances in my question. You mean that I should define It mathematically first.
– engineer100
Jan 4 at 9:00
I have answered your original question, I can't tell you why a figure you've produced doesn't show the results you expected... You've not given a Minimal, Complete, and Verifiable example to reproduce these results, or any explanation as to why they aren't as expected? This perhaps needs its own question, if the original question here about implementing a custom distance measure has been answered. Please mark my answer as accepted if this is the case.
– Wolfie
Jan 4 at 10:21
I have answered your original question, I can't tell you why a figure you've produced doesn't show the results you expected... You've not given a Minimal, Complete, and Verifiable example to reproduce these results, or any explanation as to why they aren't as expected? This perhaps needs its own question, if the original question here about implementing a custom distance measure has been answered. Please mark my answer as accepted if this is the case.
– Wolfie
Jan 4 at 10:21
add a comment |
Thanks for contributing an answer to Stack Overflow!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f54013724%2fhow-to-compute-distance-between-two-vectors-based-on-measures%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
