You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Can I ask what kind of GPU you trained it on and if you have an idea of minimum spec to run this model? I keep hitting memory errors but then I only have RTX 2070. I'm considering Titan RTX (24GB), do you think that would be enough? Would be sad to spend all that money and still not be able to use models like this!
Best wishes,
Mark
PS in case it helps anyone else - the only way I got this working with GPU in a Conda env was with:
Hi,
Thanks for the great implementation.
Can I ask what kind of GPU you trained it on and if you have an idea of minimum spec to run this model? I keep hitting memory errors but then I only have RTX 2070. I'm considering Titan RTX (24GB), do you think that would be enough? Would be sad to spend all that money and still not be able to use models like this!
Best wishes,
Mark
PS in case it helps anyone else - the only way I got this working with GPU in a Conda env was with:
conda create -n [envname] python=3.6
conda activate [envname]
conda install -c anaconda tensorflow-gpu==1.9
pip install keras==2.2.2
(and install scipy, mkl)
It seems that there are problems with conda-forge and anaconda versions of tensorflow-gpu==1.8, the above was the only thing that worked for me.
The text was updated successfully, but these errors were encountered: