・ <2017/05/28 postscript> It seems that it is better to create a virtual environment with createconda of Anaconda. So this method seems to be wrong.
I myself couldn't download the GPU-compatible version of TensorFlow and struggled quite a bit, but it has been resolved so I'll write it here for those with similar concerns.
We will proceed for PCs with NVIDIA graphics boards in a Windows environment.
First install Anaconda. (If you have already installed it, you can skip it.) Anaconda Please install Python3.6 Verision here. As for the installer, all you have to do is follow the instructions.
Launch Anaconda Prompt and type ** conda install python = 3.5.2 ** to run it. that's all.
You should now be able to downgrade your Python version from 3.6 to 3.5.2.
If the location of Anaconda Prompt is the default (C: \ Users \ username \ AppData \ Roaming \ Microsoft \ Windows \ Start Menu \ Programs \ Anaconda3 (64-bit)) Should be in.
pip install tensorflow-gpu
You should now have a GPU-enabled version of TensorFlow installed.
If you have any questions, please ask a question.
Thank you for watching so far.
Recommended Posts