Anaconda을 이용하여 ROS + Tensorflow 함께 사용하기

-- CUDA, cuDNN 버전확인
https://stackoverflow.com/questions/41714757/how-to-find-cuda-version-in-ubuntu/42122965

$ nvcc --version

cuda8.0, cudnn6.0

-- 아나콘다 python2.7 버전 인스톨
https://www.anaconda.com/download/#linux

python3.x이랑 ROS 같이 써보려고 했는데, 아직 실력이 부족해서 그런지 실패..

$ bash Anaconda2-5.3.0-Linux-x86_64.sh
$ source ~/.bashrc

$ python -V
Python 2.7.15 :: Anaconda, Inc.

$ conda create -n tf14 pip python=2.7
$ source activate tf14

-- ROS 관련 패키지 인스톨

(tf14) $ pip install --upgrade pip
(tf14) $ pip install -U rosinstall msgpack empy defusedxml netifaces


--CUDA, cuDNN, CPU/GPU을 사양에 맞춰서 tensorflow download
https://github.com/mind/wheels#mkl

(tf14) $ pip install tensorflow-1.4.0-cp27-cp27mu-linux_x86_64.whl

잘 설치가 되었는지 Hello, tensorflow 실행

(tf14) $ python
Python 2.7.15 |Anaconda, Inc.| (default, May  1 2018, 23:32:55)
[GCC 7.2.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> hello = tf.constant('hello')
>>> sess=tf.Session()
2018-10-05 06:22:34.091802: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:892] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2018-10-05 06:22:34.092232: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1030] Found device 0 with properties:
name: GeForce GTX 1070 major: 6 minor: 1 memoryClockRate(GHz): 1.645
pciBusID: 0000:01:00.0
totalMemory: 7.92GiB freeMemory: 7.23GiB
2018-10-05 06:22:34.092258: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GTX 1070, pci bus id: 0000:01:00.0, compute capability: 6.1)
>>> print(sess.run(hello))
hello

참고 동영상

TurtleBot3 49 Machine Learning tutorial 1