Some notes on the usual commands needed to make machines follow our commands...
- https://stackoverflow.com/questions/50622525/which-tensorflow-and-cuda-version-combinations-are-compatible
- https://www.tensorflow.org/install/source#tested_build_configurations
lsb_release -a
nvidia-smi
whereis -b cuda
nvcc --version
/usr/local/cuda/bin/nvcc --version
whereis -b cudnn
cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2
which python
which python3
python --version
python -c 'import tensorflow as tf; print(tf.__version__)'
pip list | grep tensorflow
pip show tensorflow
conda list | grep tensorflow
import tensorflow as tf
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
/usr/local/cuda-8.0/bin/cuda-install-samples-8.0.sh .
cd NVIDIA_CUDA-8.0_Samples/1_Utilities/deviceQuery
make && ./deviceQuery
conda create --name tf_gpu tensorflow-gpu