Issue
I have installed tensorflow-gpu 1.12 via pip. I’d like to know which CUDA, cuDNN this binary was built against. Is there a way to find this information directly from the tensorflow package? e.g.
import tensorflow as tf
tf.get_build_information()
I’ve tried searching google, SO, and the tensorflow documentation but haven’t found anything useful.
I thought it might exist as numpy implements something similar:
import numpy as np
np.show_config()
Solution
As of TensorFlow 2.4.1, We can use tensorflow.python.platform.build_info to get information on which CUDA, cuDNN the binary was built against.
>>> import tensorflow
>>> print(tensorflow.__version__)
'2.4.1'
>>> import tensorflow.python.platform.build_info as build
>>> print(build.build_info)
OrderedDict([('cpu_compiler', '/usr/bin/gcc-5'), ('cuda_compute_capabilities', ['sm_35', 'sm_50', 'sm_60', 'sm_70', 'sm_75', 'compute_80']), ('cuda_version', '11.0'), ('cudnn_version', '8'), ('is_cuda_build', True), ('is_rocm_build', False)])
The build.build_info is an OrderedDict. So to get CuDNN and CUDA versions:
>>> print(build.build_info['cuda_version'])
11.0
>>> print(build.build_info['cudnn_version'])
8
Note: As this is not a public API, things can change in future versions. In previous versions, we could do:
from tensorflow.python.platform import build_info as tf_build_info;
print(tf_build_info.cuda_version_number)
Answered By – tpk
This Answer collected from stackoverflow, is licensed under cc by-sa 2.5 , cc by-sa 3.0 and cc by-sa 4.0