ModuleNotFoundError: No module named 'mxnet.gluon.metric' #19341
-
Hi, While I am trying to run the example from mxnet/example/gluon. I meet this problem: root@server:/users/yxd429/mxnet2/mxnet/example/gluon# /usr/bin/python3 image_classification.py --dataset cifar10 --model vgg11 --epochs 1 I am trying to run the training no cpu. I install it as "pip3 install mxnet" What is the problem should be ? Best Regards, |
Beta Was this translation helpful? Give feedback.
Replies: 6 comments 1 reply
-
MXNet 2 is currently unreleased. You can track the status for beta release at #19139 You can use the v1.x branch to get MXNet 1 or compile MXNet 2 from source. We'll provide package on pypi once the beta is approved later. If you wish to help test MXNet 2, you can also install a nightly build |
Beta Was this translation helpful? Give feedback.
-
What i did is i change the source code from mxnet.gluon.metrics to
mxnet.metrics.
It works now.
thank you
Leonard Lausen <[email protected]>于2020年10月12日 周一下午7:15写道:
pip3 install mxnet installs MXNet 1 and you are executing MXNet 2 Python
code with MXNet 1, which is expected to fail.
MXNet 2 is currently unreleased. You can track the status for beta release
at #19139 <#19139>
You can use the v1.x branch to get MXNet 1 or compile MXNet 2 from source.
We'll provide package on pypi once the beta is approved later. If you wish
to help test MXNet 2, you can also install a nightly build pip install
--pre -f https://dist.mxnet.io/python/cpu mxnet
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#19341 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ANDGLBYSPPHWCTE4ZYV3MJDSKOE2FANCNFSM4SNTKD5Q>
.
--
Yuanjun Dai (he/him)
P.hd
Department of Computer and Data Sciences
Case Western Reserve University
Phone: (216)-235-8330
Office: Glennan 505
|
Beta Was this translation helpful? Give feedback.
-
Hi, Leonard
Actually, I follow the instructions from the tutorial from this link.
I just try to run the example model here.
https://mxnet.apache.org/versions/1.7/api/faq/distributed_training
All the command I am using is from this link...
However, the errors keep showing up.
What tutorial I need to follow with ?
[apache/incubator-mxnet] ModuleNotFoundError: No module named
'mxnet.gluon.metric'
Best Regards,
Johnny
…On Mon, Oct 12, 2020 at 7:15 PM Leonard Lausen ***@***.***> wrote:
pip3 install mxnet installs MXNet 1 and you are executing MXNet 2 Python
code with MXNet 1, which is expected to fail.
MXNet 2 is currently unreleased. You can track the status for beta release
at #19139 <#19139>
You can use the v1.x branch to get MXNet 1 or compile MXNet 2 from source.
We'll provide package on pypi once the beta is approved later. If you wish
to help test MXNet 2, you can also install a nightly build pip install
--pre -f https://dist.mxnet.io/python/cpu mxnet
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#19341 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ANDGLBYSPPHWCTE4ZYV3MJDSKOE2FANCNFSM4SNTKD5Q>
.
--
Yuanjun Dai (he/him)
P.hd
Department of Computer and Data Sciences
Case Western Reserve University
Phone: (216)-235-8330
Office: Glennan 505
|
Beta Was this translation helpful? Give feedback.
-
Thank you so much.
I will try.
Best Regards,
Johnny
…On Sat, Oct 17, 2020 at 11:10 AM Sheng Zha ***@***.***> wrote:
The problem with the page is that it links to master branch, which
actually should have been linked to a commit instead. i.e.
This page:
https://github.com/apache/incubator-mxnet/blame/v1.7.x/docs/static_site/src/pages/api/faq/distributed_training.md#L98
should link to:
https://github.com/apache/incubator-mxnet/blob/7efb5df7fcaaff5b1ced41b7e718e2146ed57814/example/gluon/image_classification.py
For now you can run the script from the second link
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#19341 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ANDGLB7OHJUV6PWMDYKDC2LSLGXVHANCNFSM4SNTKD5Q>
.
--
Yuanjun Dai (he/him)
P.hd
Department of Computer and Data Sciences
Case Western Reserve University
Phone: (216)-235-8330
Office: Glennan 505
|
Beta Was this translation helpful? Give feedback.
-
Hi,
Thank you for your guys' quick response !
Hope this E-mail finds you well!
I use a right image_classification.py file, however, I still have other
problems.
Problem One:
Within the tutorial for distributed training, the information within the
hostfile should be only IP address.
My master can ssh to the worker nodes using command "ssh -A user@ip_address".
However, if the hostfile only contains the ip_address without the user
name, the ssh will be denied because of the publickkey problem. However, if
the hostfile contains user@ip_address, there will be no ssh problem.
Problem Two:
When I try to run the command:
../../tools/launch.py -n 3 -s 1 -H hosts --launcher ssh --sync-dst-dir
/users/yxd429/test python image_classification.py --dataset cifar10 --model
vgg11 --epochs 1 --kvstore dist_sync
There will be a errors:
rsync: failed to set times on "/users/yxd429/test/.": Operation not
permitted (1)
How can I solve this problem ?
Best Regards,
Johnny
…--
Yuanjun Dai (he/him)
P.hd
Department of Computer and Data Sciences
Case Western Reserve University
Phone: (216)-235-8330
Office: Glennan 505
|
Beta Was this translation helpful? Give feedback.
-
Also, even though I follow all the instructions, I still got this type of
errors:
What is wrong with my implementation ?
root@server:/users/yxd429/incubator-mxnet/example/gluon#
../../tools/launch.py -n 3 -s 1 -H hosts --launcher ssh --sync-dst-dir
/users/yxd429/test python image_classification.py --dataset cifar10 --model
vgg11 --epochs 1 --kvstore dist_sync
2020-10-17 13:49:04,653 INFO rsync
/users/yxd429/incubator-mxnet/example/gluon/ -> [email protected]
[email protected] [email protected] [email protected]:
/users/yxd429/test
bash: [email protected]: command not found
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
python: can't open file 'image_classification.py': [Errno 2] No such file
or directory
bash: [email protected]: command not found
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
bash: [email protected]: command not found
bash: [email protected]: command not found
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
bash: line 0: export: `[email protected]': not a valid identifier
Best Regards,
Johnny
…On Sat, Oct 17, 2020 at 11:10 AM Sheng Zha ***@***.***> wrote:
The problem with the page is that it links to master branch, which
actually should have been linked to a commit instead. i.e.
This page:
https://github.com/apache/incubator-mxnet/blame/v1.7.x/docs/static_site/src/pages/api/faq/distributed_training.md#L98
should link to:
https://github.com/apache/incubator-mxnet/blob/7efb5df7fcaaff5b1ced41b7e718e2146ed57814/example/gluon/image_classification.py
For now you can run the script from the second link
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#19341 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ANDGLB7OHJUV6PWMDYKDC2LSLGXVHANCNFSM4SNTKD5Q>
.
--
Yuanjun Dai (he/him)
P.hd
Department of Computer and Data Sciences
Case Western Reserve University
Phone: (216)-235-8330
Office: Glennan 505
|
Beta Was this translation helpful? Give feedback.
pip3 install mxnet
installs MXNet 1 and you are executing MXNet 2 Python code with MXNet 1, which is expected to fail.MXNet 2 is currently unreleased. You can track the status for beta release at #19139
You can use the v1.x branch to get MXNet 1 or compile MXNet 2 from source. We'll provide package on pypi once the beta is approved later. If you wish to help test MXNet 2, you can also install a nightly build
pip install --pre -f https://dist.mxnet.io/python/cpu mxnet