tensorflow1.13.1+cuda10.0

上一篇帖子里win10+vs2017+cuda10.1+python3.7.2成功安装上了tensorflow的gpu版本，回想起来，有如下几点坑：

1.python学习之旅需要用到大量第三方库，推荐使用anaconda，非常方便管理库及其库的依赖，说干就干呗，于是准备下载了anaconda，先是去官网下，无奈速度很慢，就转去清华资源下载，但问题是资源很老，安装anaconda后发现python是3.7.0的，当时想的是基于现有的环境，手动升级python，

发现根本没法下载完，手动复制网址，在浏览器下载，然后手动安装

(base) C:\Windows\system32>conda install C:\Users\WIN10\Downloads\python-3.7.2-h8c8aaf0_10.tar.bz2
Preparing transaction: done
Verifying transaction: done
Executing transaction: done

(base) C:\Windows\system32>python
Python 3.7.2 (default, Feb 21 2019, 17:35:59) [MSC v.1915 64 bit (AMD64)] :: Anaconda, Inc. on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import cv2
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ModuleNotFoundError: No module named 'cv2'
>>> exit()

(base) C:\Windows\system32>pip install opencv-python
Collecting opencv-python
  Using cached https://files.pythonhosted.org/packages/cc/3c/e9b46d4ff65d4dfcca1789e32113f38d5fc1804840a4aa0bc2437ecef860/opencv_python-4.0.0.21-cp37-cp37m-win_amd64.whl
Requirement already satisfied: numpy>=1.14.5 in c:\programdata\anaconda3\lib\site-packages (from opencv-python) (1.15.4)
Installing collected packages: opencv-python
Successfully installed opencv-python-4.0.0.21

(base) C:\Windows\system32>python
Python 3.7.2 (default, Feb 21 2019, 17:35:59) [MSC v.1915 64 bit (AMD64)] :: Anaconda, Inc. on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import cv2
>>> print(cv2.__version__)
4.0.0
>>> print(cv2.__file__)
C:\ProgramData\Anaconda3\lib\site-packages\cv2\cv2.cp37-win_amd64.pyd
>>> import tensorflow
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ModuleNotFoundError: No module named 'tensorflow'
>>> pip install --ignore-installed --upgrade tensorflow-gpu
  File "<stdin>", line 1
    pip install --ignore-installed --upgrade tensorflow-gpu
              ^
SyntaxError: invalid syntax
>>> exit()

(base) C:\Windows\system32>pip install --ignore-installed --upgrade tensorflow-gpu
Collecting tensorflow-gpu
  Using cached https://files.pythonhosted.org/packages/a0/dd/8fd5f91345ef290e884343bbb947ab074af4cb73813128b692977160aeec/tensorflow_gpu-1.13.1-cp37-cp37m-win_amd64.whl
Collecting tensorboard<1.14.0,>=1.13.0 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/0f/39/bdd75b08a6fba41f098b6cb091b9e8c7a80e1b4d679a581a0ccd17b10373/tensorboard-1.13.1-py3-none-any.whl
Collecting wheel>=0.26 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/96/ba/a4702cbb6a3a485239fbe9525443446203f00771af9ac000fa3ef2788201/wheel-0.33.1-py2.py3-none-any.whl
Collecting grpcio>=1.8.6 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/9c/50/1fd6e3cb2cc0b9f88aa396020b0a831a2e8fb63a4479d1dfb4ae64b654bf/grpcio-1.19.0-cp37-cp37m-win_amd64.whl
Collecting keras-preprocessing>=1.0.5 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/c0/bf/0315ef6a9fd3fc2346e85b0ff1f5f83ca17073f2c31ac719ab2e4da0d4a3/Keras_Preprocessing-1.0.9-py2.py3-none-any.whl
Collecting termcolor>=1.1.0 (from tensorflow-gpu)
Collecting absl-py>=0.1.6 (from tensorflow-gpu)
Collecting gast>=0.2.0 (from tensorflow-gpu)
Collecting protobuf>=3.6.1 (from tensorflow-gpu)
  Downloading https://files.pythonhosted.org/packages/34/ef/f020691889031a8e1d8cb20711daa43cfe999e0768ff6903c4bf70c2eecd/protobuf-3.7.1-cp37-cp37m-win_amd64.whl (986kB)
    100% |████████████████████████████████| 993kB 3.7MB/s
Collecting keras-applications>=1.0.6 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/90/85/64c82949765cfb246bbdaf5aca2d55f400f792655927a017710a78445def/Keras_Applications-1.0.7-py2.py3-none-any.whl
Collecting numpy>=1.13.3 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/3a/3c/515afabfe4f29bfc0a67037efaf518c33d0076b32d22ba865241cee295c4/numpy-1.16.2-cp37-cp37m-win_amd64.whl
Collecting six>=1.10.0 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/73/fb/00a976f728d0d1fecfe898238ce23f502a721c0ac0ecfedb80e0d88c64e9/six-1.12.0-py2.py3-none-any.whl
Collecting astor>=0.6.0 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/35/6b/11530768cac581a12952a2aad00e1526b89d242d0b9f59534ef6e6a1752f/astor-0.7.1-py2.py3-none-any.whl
Collecting tensorflow-estimator<1.14.0rc0,>=1.13.0 (from tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/bb/48/13f49fc3fa0fdf916aa1419013bb8f2ad09674c275b4046d5ee669a46873/tensorflow_estimator-1.13.0-py2.py3-none-any.whl
Collecting markdown>=2.6.8 (from tensorboard<1.14.0,>=1.13.0->tensorflow-gpu)
  Downloading https://files.pythonhosted.org/packages/f5/e4/d8c18f2555add57ff21bf25af36d827145896a07607486cc79a2aea641af/Markdown-3.1-py2.py3-none-any.whl (87kB)
    100% |████████████████████████████████| 92kB 3.9MB/s
Collecting werkzeug>=0.11.15 (from tensorboard<1.14.0,>=1.13.0->tensorflow-gpu)
  Downloading https://files.pythonhosted.org/packages/24/4d/2fc4e872fbaaf44cc3fd5a9cd42fda7e57c031f08e28c9f35689e8b43198/Werkzeug-0.15.1-py2.py3-none-any.whl (328kB)
    100% |████████████████████████████████| 337kB 4.9MB/s
Collecting setuptools (from protobuf>=3.6.1->tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/d1/6a/4b2fcefd2ea0868810e92d519dacac1ddc64a2e53ba9e3422c3b62b378a6/setuptools-40.8.0-py2.py3-none-any.whl
Collecting h5py (from keras-applications>=1.0.6->tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/4f/1e/89aa610afce8df6fd1f12647600a05e902238587ae6375442a3164b59d51/h5py-2.9.0-cp37-cp37m-win_amd64.whl
Collecting mock>=2.0.0 (from tensorflow-estimator<1.14.0rc0,>=1.13.0->tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/e6/35/f187bdf23be87092bd0f1200d43d23076cee4d0dec109f195173fd3ebc79/mock-2.0.0-py2.py3-none-any.whl
Collecting pbr>=0.11 (from mock>=2.0.0->tensorflow-estimator<1.14.0rc0,>=1.13.0->tensorflow-gpu)
  Using cached https://files.pythonhosted.org/packages/14/09/12fe9a14237a6b7e0ba3a8d6fcf254bf4b10ec56a0185f73d651145e9222/pbr-5.1.3-py2.py3-none-any.whl
Installing collected packages: wheel, numpy, setuptools, six, protobuf, grpcio, markdown, werkzeug, absl-py, tensorboard, keras-preprocessing, termcolor, gast, h5py, keras-applications, astor, pbr, mock, tensorflow-estimator, tensorflow-gpu
Successfully installed absl-py-0.7.1 astor-0.7.1 gast-0.2.2 grpcio-1.19.0 h5py-2.9.0 keras-applications-1.0.7 keras-preprocessing-1.0.9 markdown-3.1 mock-2.0.0 numpy-1.16.2 pbr-5.1.3 protobuf-3.7.1 setuptools-40.8.0 six-1.12.0 tensorboard-1.13.1 tensorflow-estimator-1.13.0 tensorflow-gpu-1.13.1 termcolor-1.1.0 werkzeug-0.15.1 wheel-0.33.1

(base) C:\Windows\system32>python
Python 3.7.2 (default, Feb 21 2019, 17:35:59) [MSC v.1915 64 bit (AMD64)] :: Anaconda, Inc. on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\__init__.py", line 24, in <module>
    from tensorflow.python import pywrap_tensorflow  # pylint: disable=unused-import
  File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\__init__.py", line 47, in <module>
    import numpy as np
  File "C:\ProgramData\Anaconda3\lib\site-packages\numpy\__init__.py", line 142, in <module>
    from . import core
  File "C:\ProgramData\Anaconda3\lib\site-packages\numpy\core\__init__.py", line 91, in <module>
    raise ImportError(msg.format(path))
ImportError: Something is wrong with the numpy installation. While importing we detected an older version of numpy in ['C:\\ProgramData\\Anaconda3\\lib\\site-packages\\numpy']. One method of fixing this is to repeatedly uninstall numpy until none is found, then reinstall this version.
>>> import tensorflow as tf
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\__init__.py", line 24, in <module>
    from tensorflow.python import pywrap_tensorflow  # pylint: disable=unused-import
  File "C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\__init__.py", line 47, in <module>
    import numpy as np
  File "C:\ProgramData\Anaconda3\lib\site-packages\numpy\__init__.py", line 142, in <module>
    from . import core
  File "C:\ProgramData\Anaconda3\lib\site-packages\numpy\core\__init__.py", line 91, in <module>
    raise ImportError(msg.format(path))
ImportError: Something is wrong with the numpy installation. While importing we detected an older version of numpy in ['C:\\ProgramData\\Anaconda3\\lib\\site-packages\\numpy']. One method of fixing this is to repeatedly uninstall numpy until none is found, then reinstall this version.
>>> exit
Use exit() or Ctrl-Z plus Return to exit
>>> exit()

(base) C:\Windows\system32>conda list

python3.7.2安装好了，但是事情并没有那么简单，我们找到python.exe，运行

问题来了，系统不认可手动安装的python3.7.2，通过检查conda list的环境也能看出来 3.7.2未知；

conda手动安装之后肯定还有一系列操作，但还没找到合适的办法，升级python之路搁浅；

2.某一篇博文提示，tensorflow与cuda需要匹配，由于本人一直无法查看最新的tensorflow.org网站，不清楚目前2019.4.1的tensorflow与cuda的匹配情况，之前都是去官网下载最新版，tensorflow1.13.1，cuda10.1，python3.7.2，然而问题就在这儿了，经过实测，tensorflow1.13.1并不适合cuda10.1，有网友表明10.0可以，然后。。。降版本。

先是卸载nVidia的NVIDIA Nsight、cuda runtime10.1，cuda visiual studio等相关的，然后重新下载cuda_10.0.130_411.31_win10，cuDnn搭配cudnn-10.0-windows10-x64-v7.4.2.24，重新安装cuda，然后通过在anaconda官网找到最新anaconda下载链接，右键检查元素，复制网址到迅雷或其他可加速的下载器下载嗖快，我的是Anaconda3-2018.12-Windows-x86_64，链接https://repo.anaconda.com/archive/Anaconda3-2018.12-Windows-x86_64.exe，然后安装的python是3.7.1，我已没有升级到3.7.2的欲望，先能用！

由于之前安装了tensorflow-gpu，重装cuda后需要重装，先卸载，再安装，如下

C:\Windows\system32>pip uninstall tensorflow-gpu
Uninstalling tensorflow-gpu-1.13.1:
  Would remove:
    c:\programdata\anaconda3\lib\site-packages\tensorflow\include\external\cub_archive\license.txt
    c:\programdata\anaconda3\lib\site-packages\tensorflow\include\tensorflow\core\common_runtime\gpu\gpu_event_mgr.h
    c:\programdata\anaconda3\lib\site-packages\tensorflow\include\tensorflow\core\common_runtime\gpu\gpu_init.h
    c:\programdata\anaconda3\lib\site-packages\tensorflow\include\tensorflow\core\kernels\ops_util.h
    c:\programdata\anaconda3\lib\site-packages\tensorflow_gpu-1.13.1.dist-info\*
Proceed (y/n)? y
  Successfully uninstalled tensorflow-gpu-1.13.1

C:\Windows\system32>pip uninstall tensorflow-gpu
Skipping tensorflow-gpu as it is not installed.

C:\Windows\system32>python
Python 3.7.1 (default, Dec 10 2018, 22:54:23) [MSC v.1915 64 bit (AMD64)] :: Anaconda, Inc. on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: module 'tensorflow' has no attribute 'constant'
>>> b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: module 'tensorflow' has no attribute 'constant'
>>> c = tf.matmul(a, b)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: module 'tensorflow' has no attribute 'matmul'
>>>
>>> sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: module 'tensorflow' has no attribute 'Session'
>>>
>>> print sess.run(c)
  File "<stdin>", line 1
    print sess.run(c)
             ^
SyntaxError: invalid syntax
>>> exit()

C:\Windows\system32>pip3 install tensorflow-gpu
Collecting tensorflow-gpu
  Using cached https://files.pythonhosted.org/packages/a0/dd/8fd5f91345ef290e884343bbb947ab074af4cb73813128b692977160aeec/tensorflow_gpu-1.13.1-cp37-cp37m-win_amd64.whl
Requirement already satisfied: numpy>=1.13.3 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (1.16.2)
Requirement already satisfied: keras-applications>=1.0.6 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (1.0.7)
Requirement already satisfied: grpcio>=1.8.6 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (1.19.0)
Requirement already satisfied: absl-py>=0.1.6 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (0.7.1)
Requirement already satisfied: six>=1.10.0 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (1.12.0)
Requirement already satisfied: tensorflow-estimator<1.14.0rc0,>=1.13.0 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (1.13.0)
Requirement already satisfied: keras-preprocessing>=1.0.5 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (1.0.9)
Requirement already satisfied: termcolor>=1.1.0 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (1.1.0)
Requirement already satisfied: tensorboard<1.14.0,>=1.13.0 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (1.13.1)
Requirement already satisfied: wheel>=0.26 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (0.33.1)
Requirement already satisfied: protobuf>=3.6.1 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (3.7.1)
Requirement already satisfied: astor>=0.6.0 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (0.7.1)
Requirement already satisfied: gast>=0.2.0 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-gpu) (0.2.2)
Requirement already satisfied: h5py in c:\programdata\anaconda3\lib\site-packages (from keras-applications>=1.0.6->tensorflow-gpu) (2.9.0)
Requirement already satisfied: mock>=2.0.0 in c:\programdata\anaconda3\lib\site-packages (from tensorflow-estimator<1.14.0rc0,>=1.13.0->tensorflow-gpu) (2.0.0)
Requirement already satisfied: markdown>=2.6.8 in c:\programdata\anaconda3\lib\site-packages (from tensorboard<1.14.0,>=1.13.0->tensorflow-gpu) (3.1)
Requirement already satisfied: werkzeug>=0.11.15 in c:\programdata\anaconda3\lib\site-packages (from tensorboard<1.14.0,>=1.13.0->tensorflow-gpu) (0.15.1)
Requirement already satisfied: setuptools in c:\programdata\anaconda3\lib\site-packages (from protobuf>=3.6.1->tensorflow-gpu) (40.8.0)
Requirement already satisfied: pbr>=0.11 in c:\programdata\anaconda3\lib\site-packages (from mock>=2.0.0->tensorflow-estimator<1.14.0rc0,>=1.13.0->tensorflow-gpu) (5.1.3)
Installing collected packages: tensorflow-gpu
Successfully installed tensorflow-gpu-1.13.1

C:\Windows\system32>python
Python 3.7.1 (default, Dec 10 2018, 22:54:23) [MSC v.1915 64 bit (AMD64)] :: Anaconda, Inc. on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')
>>> b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')
>>> c = tf.matmul(a, b)
>>>
>>> sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
2019-04-01 08:47:16.301314: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2019-04-01 08:47:17.305673: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1433] Found device 0 with properties:
name: GeForce GTX 1060 major: 6 minor: 1 memoryClockRate(GHz): 1.6705
pciBusID: 0000:01:00.0
totalMemory: 6.00GiB freeMemory: 4.97GiB
2019-04-01 08:47:17.310680: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1512] Adding visible gpu devices: 0
2019-04-01 08:47:17.849069: I tensorflow/core/common_runtime/gpu/gpu_device.cc:984] Device interconnect StreamExecutor with strength 1 edge matrix:
2019-04-01 08:47:17.851202: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990]      0
2019-04-01 08:47:17.852714: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1003] 0:   N
2019-04-01 08:47:17.854321: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 4716 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1060, pci bus id: 0000:01:00.0, compute capability: 6.1)
Device mapping:
/job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: GeForce GTX 1060, pci bus id: 0000:01:00.0, compute capability: 6.1
2019-04-01 08:47:17.863278: I tensorflow/core/common_runtime/direct_session.cc:317] Device mapping:
/job:localhost/replica:0/task:0/device:GPU:0 -> device: 0, name: GeForce GTX 1060, pci bus id: 0000:01:00.0, compute capability: 6.1

>>>
>>> print sess.run(c)

经过反复卸载、安装，本人环境为win10,vs2017,anaconda3（附带安装python3.7.1，自己下载的3.7.2已卸载）,cuda10.0安装tensorflowGPU版OK了，其实也就这三个文件，

cmd下pip安装pip install tensorflow-gpu，pip install opencv-python  就成功了。

C:\Users\WIN10>python
Python 3.7.1 (default, Dec 10 2018, 22:54:23) [MSC v.1915 64 bit (AMD64)] :: Anaconda, Inc. on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> tf.__version__
'1.13.1'
>>> tf.__path__
['C:\\ProgramData\\Anaconda3\\lib\\site-packages\\tensorflow_estimator\\python\\estimator\\api', 'C:\\ProgramData\\Anaconda3\\lib\\site-packages\\tensorflow', 'C:\\ProgramData\\Anaconda3\\lib\\site-packages\\tensorflow\\_api\\v1']
>>>

本人经验总结，不要轻易相信别人，实践出真理。