Matrix Subtract like Matrix Multiplication in tensorflow
This is my first post I usually found all my answers in the archives, but having a hard time with this one, thanks for the help!
I have two matrix A and B. Performing a matrix multiplication operation is trivial using tf.matmult. But I want to do matrix subtract similar to how matrix multiplication works. Eg if I have.
A = tf.constant([[1, 1, 1, 2, 3, 1],[1,2,3,4,5,6],[4,3,2,1,6,5]])
B = tf.constant([[1,3,1],[2,1,1]])
#B*A
X = tf.matmult(B,A)
>>>X = [[8,10,12,15,24,24],[7,7,7,9,17,13]]
What I want to do is do a similar operation like matmult, but instead of multiply I want subtract and square. Eg...
for x11, where the subscript 11 is row 1, column 1 of matrix X.
= (-b11 + a11)2 + (-b12 + a21)2 + (-b13 + a31)2
and
x12 = (-b11 + a12)2 + (-b12 + a22)2 + (-b13 + a32)2
and so on similar to how matrix multiplication works.
So if we take matrix A and B above and perform the operation described above (call it matmultsubtract), we get,
tf.matmultsubtract(B,A) gives:
[[(-1+1)2+(-3+1)2+(-1+4)2, (-1+1)2+(-3+2)2+(-1+3)2,...],
[(-2+1)2+(-1+1)2+(-1+4)2, (-2+1)2+(-1+2)2+(-1+3)2, ...]]
This isn't that hard if working with numpy arrays (you can use two nested for loops) by iterating manually rather than np.matmult, but tensorflow has a problem with for loops and I'm not sure how to do it.
Thanks for the help.
python tensorflow matrix matrix-multiplication
asked Dec 28 '18 at 0:46
Harp
84
add a comment |
This is my first post I usually found all my answers in the archives, but having a hard time with this one, thanks for the help!
I have two matrix A and B. Performing a matrix multiplication operation is trivial using tf.matmult. But I want to do matrix subtract similar to how matrix multiplication works. Eg if I have.
A = tf.constant([[1, 1, 1, 2, 3, 1],[1,2,3,4,5,6],[4,3,2,1,6,5]])
B = tf.constant([[1,3,1],[2,1,1]])
#B*A
X = tf.matmult(B,A)
>>>X = [[8,10,12,15,24,24],[7,7,7,9,17,13]]
What I want to do is do a similar operation like matmult, but instead of multiply I want subtract and square. Eg...
for x11, where the subscript 11 is row 1, column 1 of matrix X.
= (-b11 + a11)2 + (-b12 + a21)2 + (-b13 + a31)2
and
x12 = (-b11 + a12)2 + (-b12 + a22)2 + (-b13 + a32)2
and so on similar to how matrix multiplication works.
So if we take matrix A and B above and perform the operation described above (call it matmultsubtract), we get,
tf.matmultsubtract(B,A) gives:
[[(-1+1)2+(-3+1)2+(-1+4)2, (-1+1)2+(-3+2)2+(-1+3)2,...],
[(-2+1)2+(-1+1)2+(-1+4)2, (-2+1)2+(-1+2)2+(-1+3)2, ...]]
This isn't that hard if working with numpy arrays (you can use two nested for loops) by iterating manually rather than np.matmult, but tensorflow has a problem with for loops and I'm not sure how to do it.
Thanks for the help.
python tensorflow matrix matrix-multiplication
asked Dec 28 '18 at 0:46
Harp
84
If you don't mind me asking, why do you need this? I have never seen someone asking to do that... Matrix multiplication is the way it is because that's how we would compose linear operators, but why would you need a similar structure for [squared] subtraction?
– Fred
Dec 28 '18 at 1:07
I imagine it has something to do with MSE. With that in mind I'm wondering what you want to do because there might be a better way of doing it
– Fred
Dec 28 '18 at 1:08
You can do that usingtf.while_loop, it's just a bit complex-ish. Let me give it a go
– Fred
Dec 28 '18 at 1:11
add a comment |
This is my first post I usually found all my answers in the archives, but having a hard time with this one, thanks for the help!
I have two matrix A and B. Performing a matrix multiplication operation is trivial using tf.matmult. But I want to do matrix subtract similar to how matrix multiplication works. Eg if I have.
A = tf.constant([[1, 1, 1, 2, 3, 1],[1,2,3,4,5,6],[4,3,2,1,6,5]])
B = tf.constant([[1,3,1],[2,1,1]])
#B*A
X = tf.matmult(B,A)
>>>X = [[8,10,12,15,24,24],[7,7,7,9,17,13]]
What I want to do is do a similar operation like matmult, but instead of multiply I want subtract and square. Eg...
for x11, where the subscript 11 is row 1, column 1 of matrix X.
= (-b11 + a11)2 + (-b12 + a21)2 + (-b13 + a31)2
and
x12 = (-b11 + a12)2 + (-b12 + a22)2 + (-b13 + a32)2
and so on similar to how matrix multiplication works.
So if we take matrix A and B above and perform the operation described above (call it matmultsubtract), we get,
tf.matmultsubtract(B,A) gives:
[[(-1+1)2+(-3+1)2+(-1+4)2, (-1+1)2+(-3+2)2+(-1+3)2,...],
[(-2+1)2+(-1+1)2+(-1+4)2, (-2+1)2+(-1+2)2+(-1+3)2, ...]]
This isn't that hard if working with numpy arrays (you can use two nested for loops) by iterating manually rather than np.matmult, but tensorflow has a problem with for loops and I'm not sure how to do it.
Thanks for the help.
python tensorflow matrix matrix-multiplication
asked Dec 28 '18 at 0:46
Harp
84
This is my first post I usually found all my answers in the archives, but having a hard time with this one, thanks for the help!
I have two matrix A and B. Performing a matrix multiplication operation is trivial using tf.matmult. But I want to do matrix subtract similar to how matrix multiplication works. Eg if I have.
A = tf.constant([[1, 1, 1, 2, 3, 1],[1,2,3,4,5,6],[4,3,2,1,6,5]])
B = tf.constant([[1,3,1],[2,1,1]])
#B*A
X = tf.matmult(B,A)
>>>X = [[8,10,12,15,24,24],[7,7,7,9,17,13]]
What I want to do is do a similar operation like matmult, but instead of multiply I want subtract and square. Eg...
for x11, where the subscript 11 is row 1, column 1 of matrix X.
= (-b11 + a11)2 + (-b12 + a21)2 + (-b13 + a31)2
and
x12 = (-b11 + a12)2 + (-b12 + a22)2 + (-b13 + a32)2
and so on similar to how matrix multiplication works.
So if we take matrix A and B above and perform the operation described above (call it matmultsubtract), we get,
tf.matmultsubtract(B,A) gives:
[[(-1+1)2+(-3+1)2+(-1+4)2, (-1+1)2+(-3+2)2+(-1+3)2,...],
[(-2+1)2+(-1+1)2+(-1+4)2, (-2+1)2+(-1+2)2+(-1+3)2, ...]]
This isn't that hard if working with numpy arrays (you can use two nested for loops) by iterating manually rather than np.matmult, but tensorflow has a problem with for loops and I'm not sure how to do it.
Thanks for the help.
python tensorflow matrix matrix-multiplication
python tensorflow matrix matrix-multiplication
asked Dec 28 '18 at 0:46
Harp
84
asked Dec 28 '18 at 0:46
Harp
84
edited Dec 28 '18 at 1:15
asked Dec 28 '18 at 0:46
Harp
84
asked Dec 28 '18 at 0:46
Harp
84
asked Dec 28 '18 at 0:46
Harp
84
84
If you don't mind me asking, why do you need this? I have never seen someone asking to do that... Matrix multiplication is the way it is because that's how we would compose linear operators, but why would you need a similar structure for [squared] subtraction?
– Fred
Dec 28 '18 at 1:07
I imagine it has something to do with MSE. With that in mind I'm wondering what you want to do because there might be a better way of doing it
– Fred
Dec 28 '18 at 1:08
You can do that usingtf.while_loop, it's just a bit complex-ish. Let me give it a go
– Fred
Dec 28 '18 at 1:11
add a comment |
If you don't mind me asking, why do you need this? I have never seen someone asking to do that... Matrix multiplication is the way it is because that's how we would compose linear operators, but why would you need a similar structure for [squared] subtraction?
– Fred
Dec 28 '18 at 1:07
I imagine it has something to do with MSE. With that in mind I'm wondering what you want to do because there might be a better way of doing it
– Fred
Dec 28 '18 at 1:08
You can do that usingtf.while_loop, it's just a bit complex-ish. Let me give it a go
– Fred
Dec 28 '18 at 1:11
If you don't mind me asking, why do you need this? I have never seen someone asking to do that... Matrix multiplication is the way it is because that's how we would compose linear operators, but why would you need a similar structure for [squared] subtraction?
– Fred
Dec 28 '18 at 1:07
If you don't mind me asking, why do you need this? I have never seen someone asking to do that... Matrix multiplication is the way it is because that's how we would compose linear operators, but why would you need a similar structure for [squared] subtraction?
– Fred
Dec 28 '18 at 1:07
I imagine it has something to do with MSE. With that in mind I'm wondering what you want to do because there might be a better way of doing it
– Fred
Dec 28 '18 at 1:08
I imagine it has something to do with MSE. With that in mind I'm wondering what you want to do because there might be a better way of doing it
– Fred
Dec 28 '18 at 1:08
You can do that using
tf.while_loop, it's just a bit complex-ish. Let me give it a go– Fred
Dec 28 '18 at 1:11
You can do that using
tf.while_loop, it's just a bit complex-ish. Let me give it a go– Fred
Dec 28 '18 at 1:11
add a comment |
1 Answer
1
active
oldest
votes
Trying a vectorization operation that may not be taken as matrix subtract.
# shape=(2,3,6)
B_new = tf.tile(tf.expand_dims(B,axis=-1),multiples=[1,1,A.shape[1]])
# shape=(2,3,6)
A_new = tf.tile(tf.expand_dims(A,axis=0),multiples=[B.shape[0],1,1])
# shape=(2,6)
result = tf.reduce_sum(tf.square(A_new - B_new),axis=1)
with tf.Session() as sess:
print(sess.run(result))
[[13 5 1 2 33 25]
[10 6 6 9 42 42]]
answered Dec 28 '18 at 3:57
giser_yugang
1,3941318
add a comment |
Your Answer
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%2f53952470%2fmatrix-subtract-like-matrix-multiplication-in-tensorflow%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
Trying a vectorization operation that may not be taken as matrix subtract.
# shape=(2,3,6)
B_new = tf.tile(tf.expand_dims(B,axis=-1),multiples=[1,1,A.shape[1]])
# shape=(2,3,6)
A_new = tf.tile(tf.expand_dims(A,axis=0),multiples=[B.shape[0],1,1])
# shape=(2,6)
result = tf.reduce_sum(tf.square(A_new - B_new),axis=1)
with tf.Session() as sess:
print(sess.run(result))
[[13 5 1 2 33 25]
[10 6 6 9 42 42]]
answered Dec 28 '18 at 3:57
giser_yugang
1,3941318
add a comment |
Trying a vectorization operation that may not be taken as matrix subtract.
# shape=(2,3,6)
B_new = tf.tile(tf.expand_dims(B,axis=-1),multiples=[1,1,A.shape[1]])
# shape=(2,3,6)
A_new = tf.tile(tf.expand_dims(A,axis=0),multiples=[B.shape[0],1,1])
# shape=(2,6)
result = tf.reduce_sum(tf.square(A_new - B_new),axis=1)
with tf.Session() as sess:
print(sess.run(result))
[[13 5 1 2 33 25]
[10 6 6 9 42 42]]
answered Dec 28 '18 at 3:57
giser_yugang
1,3941318
add a comment |
Trying a vectorization operation that may not be taken as matrix subtract.
# shape=(2,3,6)
B_new = tf.tile(tf.expand_dims(B,axis=-1),multiples=[1,1,A.shape[1]])
# shape=(2,3,6)
A_new = tf.tile(tf.expand_dims(A,axis=0),multiples=[B.shape[0],1,1])
# shape=(2,6)
result = tf.reduce_sum(tf.square(A_new - B_new),axis=1)
with tf.Session() as sess:
print(sess.run(result))
[[13 5 1 2 33 25]
[10 6 6 9 42 42]]
answered Dec 28 '18 at 3:57
giser_yugang
1,3941318
Trying a vectorization operation that may not be taken as matrix subtract.
# shape=(2,3,6)
B_new = tf.tile(tf.expand_dims(B,axis=-1),multiples=[1,1,A.shape[1]])
# shape=(2,3,6)
A_new = tf.tile(tf.expand_dims(A,axis=0),multiples=[B.shape[0],1,1])
# shape=(2,6)
result = tf.reduce_sum(tf.square(A_new - B_new),axis=1)
with tf.Session() as sess:
print(sess.run(result))
[[13 5 1 2 33 25]
[10 6 6 9 42 42]]
answered Dec 28 '18 at 3:57
giser_yugang
1,3941318
answered Dec 28 '18 at 3:57
giser_yugang
1,3941318
answered Dec 28 '18 at 3:57
giser_yugang
1,3941318
answered Dec 28 '18 at 3:57
giser_yugang
1,3941318
1,3941318
add a comment |
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.
Some of your past answers have not been well-received, and you're in danger of being blocked from answering.
Please pay close attention to the following guidance:
- 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%2f53952470%2fmatrix-subtract-like-matrix-multiplication-in-tensorflow%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
If you don't mind me asking, why do you need this? I have never seen someone asking to do that... Matrix multiplication is the way it is because that's how we would compose linear operators, but why would you need a similar structure for [squared] subtraction?
– Fred
Dec 28 '18 at 1:07
I imagine it has something to do with MSE. With that in mind I'm wondering what you want to do because there might be a better way of doing it
– Fred
Dec 28 '18 at 1:08
You can do that using
tf.while_loop, it's just a bit complex-ish. Let me give it a go– Fred
Dec 28 '18 at 1:11