I'm using python 2.7 and Keras 2.1.4.
Out of the following versions (cuda/10.0 ,10.1, 10.2, 11.0, 11.1, 11.1.0, 11.1.1, 11.2, 11.2.2, 9.0, 9.1, 9.2) which version can I use for Tensorflow 1.8.0?
Is there any specific documentation? where I could read more about these.
You can use this table to check compatibilities. For version 1.8.0, it is cuda/9.
Related
in opensuse leap15.2, how can I upgrade my gnuplot (current version 5.2) to version 5.4 or 5.5. I want to install directly via yast the rpm package, so do not compile from source myself?
I would like to use tensorflow with ROS and ROS 1 does only support python 2.7 at the moment.
Is there a way to install tensorflow 1 or 2 for python 2.7?
Thanks
According to this older post: How to install Tensorflow on Python 2.7 on Windows?
Tensorflow does not support python2.7 anymore.
This is the system requirement for Tensorflow available: https://www.tensorflow.org/install/pip?hl=en :
System requirements
Python 3.5–3.8
Python 3.8 support requires TensorFlow 2.2 or later.
pip 19.0 or later (requires manylinux2010 support)
Ubuntu 16.04 or later (64-bit)
macOS 10.12.6 (Sierra) or later (64-bit) (no GPU support)
Windows 7 or later (64-bit)
Microsoft Visual C++ Redistributable for Visual Studio 2015, 2017 and 2019
Raspbian 9.0 or later
GPU support requires a CUDA®-enabled card (Ubuntu and Windows)
Here you have the reference to the C++ reference: https://www.tensorflow.org/api_docs/cc
The easiest way of using tensorflow is to download anaconda and then download the packages of anaconda!
On my package manager for canopy, every time I try to download opencv it downgrades several other important packages. I am then not able to upgrade those same packages or run my code. How can I download opencv without downgrading my other packages?
You haven't provided any version or platform information. But perhaps you are using an old Canopy version (current is 2.1.9), or perhaps you are using the subscriber-only "full" installer, which is only intended for airgapped or other non-updateable systems. Otherwise, the currently supported version of opencv is 3.2.0 (build 3.2.0-4) which depends on numpy 1.13.3, which is the currently supported version of numpy.
What are the minimum compute capability requirements for:
CUDA 5.5?
CUDA 6.0?
CUDA 6.5?
CUDA VERSION Min CC Deprecated CC Default CC Max CC
5.5 (and prior) 1.0 N/A 1.0
6.0 1.0 1.0 1.0
6.5 1.1 1.x 2.0
7.x 2.0 N/A 2.0
8.0 2.0 2.x 2.0 6.2
9.x 3.0 N/A 3.0 7.0
10.x 3.0 N/A 3.0 7.5 (3.0 deprecated in 10.2)
11.x 3.5 3.x 5.2 8.6 (11.0:8.0, 11.1:8.6, 11.8:9.0)
(CUDA 11.5 still "supports" cc3.5 devices; the R495 driver in CUDA 11.5 installer does not)
12.0 5.0 N/A 5.2 9.0
Min CC = minimum compute capability that can be specified to nvcc (for that toolkit version)
Deprecated CC = If you specify this CC, you will get a deprecation message, but compile should still proceed.
Default CC = The architecture that will be targetted if no -arch or -gencode switches are used
Max CC = The highest compute capability you can specify on the compile command line via arch switches (compute_XY, sm_XY)
Actually for CUDA 10.0:
with Tensorflow: The minimum required Cuda capability is 3.5.
with Pytorch: Found GPU0 GeForce GTX 650 which is of cuda capability 3.0.
PyTorch no longer supports this GPU because it is too old.
i want to install latest version of GCC on RHEL 6.5 version.There is a old version of GCC already installed(GCC version 4.4.7 20120313).I want a way such that i need to install GCC in my private folder and only i can use this to compile code.other users who compile should be based on older version.