-
Notifications
You must be signed in to change notification settings - Fork 32
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
what are the modification i should do to change this problem to solve the multiclass problem(more than 2 class).For Example cat,dog,horse,cow... #2
Comments
this other repository i have is probably easier to use and runs faster. Anyway, to do more than 2 classes, you'd need to change the classifier.py script, and do the labelling accordingly for your different classes. You could also check out the retrain code on the tensorflow github. To use it for multiclass problems you'll have to just put your classes into different folders and run the retrain.py code. The code is rather long though, so it might take a while to understand. |
Can you please modify the classifier.py script for multiclass classification |
If your dataset of images is named like cat001, cat002 and so on then it'll be easy, you could do the same as I did, which was to just attach labels based on the image filenames. Or else you'd have to find some other way to do the labelling, depending on your dataset. |
Thank you very much for your quick reply.Actually I am new in Tensorflow.
Here I am attaching my own dataset. Can you please use this dataset to
solve classification problem.Thank you in advanced
main.tar.gz
<https://drive.google.com/file/d/0B2zF7Rk6hBjKbFd2V29NOXpnOVE/view?usp=drive_web>
Sincerely
Indrajit Kalita
M.Tech(IT)
Tezpur University
…On Fri, Mar 10, 2017 at 7:40 PM, See Jie Xun ***@***.***> wrote:
If your dataset of images is named like cat001, cat002 and so on then
it'll be easy, you could do the same as I did, which was to just attach
labels based on the image filenames. Or else you'd have to find some other
way to do the labelling, depending on your dataset.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#2 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AKLNF4HvkeE-3IF6AbGqg6lAITjVfpigks5rkVnWgaJpZM4MXxzH>
.
--
Sincerely
Indrajit Kalita
M.Tech in IT
Tezpur Central University
|
Apologies I'm a little busy but for now you could look at the create_image_lists function in the tensorflow retrain script. It deals with different classes being stored inrespective different folders. Essentially it uses the gfile module which would probably be what you need. I'll let you know if I'm able to do it. Also, do use my other repository here as it'll be much faster in producing the bottleneck files. |
Thank You very much.That above mention script works for me.
…On Sat, Mar 11, 2017 at 10:31 AM, See Jie Xun ***@***.***> wrote:
Apologies I'm a little busy but for now you could look at the
create_image_lists function in the tensorflow retrain script
<https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/image_retraining/retrain.py>.
It deals with different classes being stored inrespective different
folders. Essentially it uses the gfile module which would probably be what
you need.
I'll let you know if I'm able to do it. Also, do use my other repository
here <https://github.com/jiexunsee/New-Simple-Inception-Transfer-Learning>
as it'll be much faster in producing the bottleneck files.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#2 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AKLNFxnAwS8xeJcTsev540IHyAeKElbfks5rkirGgaJpZM4MXxzH>
.
--
Sincerely
Indrajit Kalita
M.Tech in IT
Tezpur Central University
|
you're welcome :) |
No description provided.
The text was updated successfully, but these errors were encountered: