Can this model run under tensorflow 1.13 and torch 1.0.1? #3
Comments
|
I'm not sure, it has only been tested on tf 0.12.1 and pytorch 0.4.1 @sangmichaelxie do you know the answer? |
|
Okay, I will try. Btw, can this model be accelerated using GPU |
|
The code doesn't work for newer versions of TF right now. It does use GPU.
…On Thu, Apr 4, 2019 at 6:36 PM Kuan Lu(Frank) ***@***.***> wrote:
Okay, I will try. Btw, can this model be accelerated using GPU
—
You are receiving this because you were mentioned
AB93
.
Reply to this email directly, view it on GitHub
<#3 (comment)>, or mute
the thread
<https://github.com/notifications/unsubscribe-auth/AHidSrmKA_pPMMY3kxfbxASpzAuSj8QQks5vdqiigaJpZM4cd_WI>
.
|
Hi, many thanks for providing code for all methods on semi supervised regression task. I've run the code and have several questions. 1. Even I enable the use_gpu flag, it seems that there are no any GPUs used. 2. The results look different from those shown in the paper (e.g., 2.04 for ssdkl using 100 labeled samples). 3. Is 'semisup' the ssdkl method? 4. Do you have the pytorch version for this method? |
|
You should check your LD_LIBRARY_PATH and associated environment variables
that cuda is correctly linked, and that tensorflow-gpu is installed
(without tensorflow also installed). For some of the methods (like label
prop) we didn't provide a GPU implementation. When we ran the experiments,
the results were stable - make sure you're using the provided config.yaml
and print_results.py script. Also notice that we report median "Relative to
DKL" results in the paper. Yes, semisup is the SSDKL method. We don't have
pytorch code.
…On Tue, Jul 16, 2019 at 4:18 AM WeiHongLee ***@***.***> wrote:
The code doesn't work for newer versions of TF right now. It does use GPU.
… <#m_7983273833815760253_>
Hi, many thanks for providing code for all methods on semi supervised
regression task. I've run the code and have several questions. 1. Even I
enable the use_gpu flag, it seems that there are no any GPUs used. 2. The
results look different from those shown in the paper (e.g., 2.04 for ssdkl
using 100 labeled samples). 3. Is 'semisup' the ssdkl method? 4. Do you
have the pytorch version for this method?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#3?email_source=notifications&email_token=AB4J2STNQ7XKRIAWQVBUQPTP7WU7JA5CNFSM4HDX6WEKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD2AQ57A#issuecomment-511774460>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AB4J2SRKQSZUKMXIF2XGKTDP7WU7JANCNFSM4HDX6WEA>
.
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Can this model run under tensorflow 1.13 and torch 1.0.1?
The text was updated successfully, but these errors were encountered: