tensorflow-windows-wheel
tensorflow-windows-wheel copied to clipboard
TensorFlow 2.0.0 without AVX
Hello, In first part, thanks for this project :) If I I understood well, TF 2.0.0 without AVX instruction is in folder "/master/2.0.0/py37/CPU/sse2/tensorflow-2.0.0-cp37-cp37m-win_amd64.whl"? If yes, I have another problem. I tried install it with "pip install tensorflow-2.0.0-cp37-cp37m-win_amd64.whl" from download folder, but I have still have error: "tensorflow-2.0.0-cp37-cp37m-win_amd64.whl is not supported wheel on this platform". Can you please help me?
Thanks
@NaniSk
Please don't change the filename and check your python is 64-bit or not.
Hi, thanks for fast answer. Im trying build it via Docker. When I tried this via Python, I didnt have problem. Did you try this in Docker too? :)
If you are using linux container on windows, it is linux environment rather than windows.
Hi, thanks :) Do you have repo or can you recommend me some TF Linux without AVX?
https://github.com/yaroslavvb/tensorflow-community-wheels/issues https://github.com/lakshayg/tensorflow-build
hey I'm trying to download tensorflow 2 using >pip install tensorflow-2.0.0-cp37-cp37m-win_amd64.whl but it shows it is not a supported wheel on this platform I'm using 32 bit python 3.6.5 can u recommend me which file should I download to install tensorflow 2 thank you in advance
@pratik-99 https://github.com/fo40225/tensorflow-windows-wheel/tree/master/1.10.0/py36/CPU
@fo40225 I feel this is tensorflow 1.0 version, but the problem I am facing while loading my model on local system is that I am getting unknown layer: name error which I feel is because I am not using tensorflow 2 can you help me in any so that I could run tensorflow 2. I downloaded it but it gives me DLL import error
Tensorflow 1.10 is my last build supports 32-bit python.
Yes, I upgraded to 64 bit python 3.7.9 and downloaded tensorflow 2.0 but it shows DLL import error
Please check: CPU has AVX support or not install the newest msvc redist 2015-2019 exactly match python major version 3.X, use the newest minor version update if you have nvidia GPU, use the exactly match cuda version(include major and minor version), exactly match the cudnn major version and the same or newer minor version
You should post the full error message.