Not 100% certain what you have going on but in short no Tensorflow does not require a GPU and you shouldn’t have to build it from source unless you just feel like it.
One source stated same as with Nvidia GPU. Tensor. Flow doesn’t need CUDA to work, it can perform all operations using CPU (or TPU). If you want to work with non-Nvidia GPU, TF doesn’t have support for Open. CL yet, there are some experimental in-progress attempts to add it, but not by Google team.
Having both tensorflow and tensorflow-gpu installed can be confusing, as tensorflow will not use a GPU in any circumstance. My first suggestion would be to install CUDA 9.0, or make sure the CUDA version you are using matches the TF version you are using (i. e. what the TF version expects).
Can TensorFlow run on AMD GPU?
Yes it is possible to run tensorflow on AMD GPU’s but it would be one heck of a problem. As tensorflow uses CUDA which is proprietary it can’t run on AMD GPU’s so you need to use OPENCL for that and tensorflow isn’t written in that.
Does TensorFlow need CUDA to work?
, tensor Flow doesn’t need CUDA to work, it can perform all operations using CPU (or TPU). If you want to work with non-Nvidia GPU, TF doesn’t have support for Open. CL yet, there are some experimental in-progress attempts to add it, but not by Google team.
What version of Python does TensorFlow run on?
, tensor Flow version (use command below): 1.9 Python version: 3.6.3 I am working on a Win10 machine, with python 3.6.3 and using tensorflow 1.9, pip 18.0. I did not provide an option to install tensorflow with gpu, (i. e.), according to this and did not provide option for using GPU.
Just select the correct version (in this case, cp38 hints python 3.8 – moreover, Tensorflow 2.2.0 is used, the current version as of Jul 12 ’20). Bonus points for using a venv like explained eg in this answer.
How to use only the CPU in TF?
In your case, to use only the CPU, you can invoke the function with an empty list : For completeness, if you want to avoid that the runtime initialization will allocate all memory on the device, you can use tf., and config., and experimental., and set_memory_growth .