TensorFlow on MacOS

前篇配置了 Caffe 和 Digits,还有一个框架就是 Google 开发的 TensorFlow。

配置简单,简略配置下:

CUDA、CuDNN 环境
选择部署系统
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
# Ubuntu/Linux 64-bit, CPU only, Python 2.7
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0rc0-cp27-none-linux_x86_64.whl

# Ubuntu/Linux 64-bit, GPU enabled, Python 2.7
# Requires CUDA toolkit 7.5 and CuDNN v4\. For other versions, see "Install from sources" below.
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0rc0-cp27-none-linux_x86_64.whl

# Mac OS X, CPU only, Python 2.7:
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0rc0-py2-none-any.whl

# Mac OS X, GPU enabled, Python 2.7:
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0rc0-py2-none-any.whl

# Ubuntu/Linux 64-bit, CPU only, Python 3.4
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0rc0-cp34-cp34m-linux_x86_64.whl

# Ubuntu/Linux 64-bit, GPU enabled, Python 3.4
# Requires CUDA toolkit 7.5 and CuDNN v4\. For other versions, see "Install from sources" below.
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0rc0-cp34-cp34m-linux_x86_64.whl

# Ubuntu/Linux 64-bit, CPU only, Python 3.5
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/cpu/tensorflow-0.10.0rc0-cp35-cp35m-linux_x86_64.whl

# Ubuntu/Linux 64-bit, GPU enabled, Python 3.5
# Requires CUDA toolkit 7.5 and CuDNN v4\. For other versions, see "Install from sources" below.
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow-0.10.0rc0-cp35-cp35m-linux_x86_64.whl

# Mac OS X, CPU only, Python 3.4 or 3.5:
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.10.0rc0-py3-none-any.whl

# Mac OS X, GPU enabled, Python 3.4 or 3.5:
export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/gpu/tensorflow-0.10.0rc0-py3-none-any.whl
pip 部署
Python 2
1
sudo pip install --upgrade $TF_BINARY_URL
Python 3
1
sudo pip3 install --upgrade $TF_BINARY_URL
Env 配置
1
2
export CUDA_HOME=/usr/local/cuda
export LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/local/cuda/lib:/usr/local/cuda/extras/CUPTI/lib"
测试(Python 2.x)
1
2
3
4
5
6
7
8
9
import tensorflow as tf
hello = tf.constant('Hello, TensorFlow!')
sess = tf.Session()
print(sess.run(hello))
# Hello, TensorFlow!
a = tf.constant(10)
b = tf.constant(32)
print(sess.run(a + b))
# 42
补充
  • 遇到 Couldn’t open CUDA library libcuda.1.dylib. 或 failed to find libcuda.so on this system: Failed precondition: could not dlopen DSO: libcuda.1.dylib; dlerror: dlopen(libcuda.1.dylib, 5): image not found 问题:
    QQ20160810-2
    请确认 LD_LIBRARY_PATH 已配置,并且 libcuda.1.dylib 存在,我这里的情况是 libcuda.dylib 存在,于是我就软连接到 .1.dylib 上,运行成功。
    1
    2
    cd /usr/local/cuda/lib
    ln -s libcuda.dylib libcuda.1.dylib
    QQ20160810-3
    QQ20160810-0