blog abcd1992719g

import numpy as np
import tensorflow as tf

# Model parameters
W = tf.Variable([.3], dtype=tf.float32)
b = tf.Variable([-.3], dtype=tf.float32)
# Model input and output
x = tf.placeholder(tf.float32)
linear_model = W * x + b
y = tf.placeholder(tf.float32)
# loss
loss = tf.reduce_sum(tf.square(linear_model - y)) # sum of the squares
# optimizer
optimizer = tf.train.GradientDescentOptimizer(0.01)
train = optimizer.minimize(loss)
# training data
x_train = [1, 2, 3, 4]
y_train = [0, -1, -2, -3]
# training loop
init = tf.global_variables_initializer()
sess = tf.Session()
sess.run(init) # reset values to wrong
for i in range(1000):
	sess.run(train, {x:x_train, y:y_train})

# evaluate training accuracy
curr_W, curr_b, curr_loss = sess.run([W, b, loss], {x:x_train, y:y_train})
print("W: %s b: %s loss: %s"%(curr_W, curr_b, curr_loss))

import numpy as np

import tensorflow as tf

# Model parameters

W = tf.Variable([.3], dtype=tf.float32)

b = tf.Variable([-.3], dtype=tf.float32)

# Model input and output

x = tf.placeholder(tf.float32)

linear_model = W * x + b

y = tf.placeholder(tf.float32)

# loss

loss = tf.reduce_sum(tf.square(linear_model - y)) # sum of the squares

# optimizer

optimizer = tf.train.GradientDescentOptimizer(0.01)

train = optimizer.minimize(loss)

# training data

x_train = [1, 2, 3, 4]

y_train = [0, -1, -2, -3]

# training loop

init = tf.global_variables_initializer()

sess = tf.Session()

sess.run(init) # reset values to wrong

for i in range(1000):

sess.run(train, {x:x_train, y:y_train})

# evaluate training accuracy

curr_W, curr_b, curr_loss = sess.run([W, b, loss], {x:x_train, y:y_train})

print("W: %s b: %s loss: %s"%(curr_W, curr_b, curr_loss))

运行：

allen@T420:~/tensorflow$ python 001_liner_gradient_descent.py 
2017-07-03 15:42:22.655446: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
2017-07-03 15:42:22.655486: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2017-07-03 15:42:22.655495: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
W: [-0.9999969] b: [ 0.99999082] loss: 5.69997e-11
allen@T420:~/tensorflow$ export TF_CPP_MIN_LOG_LEVEL=2
allen@T420:~/tensorflow$ python 001_liner_gradient_descent.py 
W: [-0.9999969] b: [ 0.99999082] loss: 5.69997e-11

allen@T420:~/tensorflow$ python 001_liner_gradient_descent.py

2017-07-03 15:42:22.655446: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.

2017-07-03 15:42:22.655486: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.

2017-07-03 15:42:22.655495: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.

W: [-0.9999969] b: [ 0.99999082] loss: 5.69997e-11

allen@T420:~/tensorflow$ export TF_CPP_MIN_LOG_LEVEL=2

allen@T420:~/tensorflow$ python 001_liner_gradient_descent.py

W: [-0.9999969] b: [ 0.99999082] loss: 5.69997e-11

tensorflow 默认编译没有使用SSE4.1指令，可以自己重新编译来使用这些指令，性能有提升。

没有大量运算时，简单一点处理就是在shell里运行下面命令使报错信息不显示。

export TF_CPP_MIN_LOG_LEVEL=2

1	export TF_CPP_MIN_LOG_LEVEL=2

后面有性能要求时可重新编译：

https://github.com/tensorflow/tensorflow/issues/7778

https://github.com/tensorflow/tensorflow/issues/8037

http://www.linuxdiyf.com/linux/30128.html

tensorflow

Posted on 2017年7月2日, updated on 2017年7月3日 by allenmo

https://www.tensorflow.org/get_started/get_started

https://morvanzhou.github.io/tutorials/machine-learning/tensorflow/

https://www.leiphone.com/news/201606/ORlQ7uK3TIW8xVGF.html

http://www.tensorfly.cn/tfdoc/get_started/introduction.html

https://www.zhihu.com/question/49909565

http://wiki.jikexueyuan.com/project/tensorflow-zh/

TensorFlow深度学习，一篇文章就够了

升级（1.2.1）：

allen@T420:~$ sudo pip install --upgrade tensorflow
[sudo] password for allen: 
The directory '/home/allen/.cache/pip/http' or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.
The directory '/home/allen/.cache/pip' or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.
Collecting tensorflow
  Downloading tensorflow-1.2.1-cp27-cp27mu-manylinux1_x86_64.whl (34.5MB)
    100% |████████████████████████████████| 34.5MB 39kB/s 
Requirement already up-to-date: mock>=2.0.0 in ./.local/lib/python2.7/site-packages (from tensorflow)
Collecting numpy>=1.11.0 (from tensorflow)
  Downloading numpy-1.13.0-cp27-cp27mu-manylinux1_x86_64.whl (16.6MB)
    100% |████████████████████████████████| 16.6MB 60kB/s 
Requirement already up-to-date: bleach==1.5.0 in ./.local/lib/python2.7/site-packages (from tensorflow)
Collecting markdown>=2.6.8 (from tensorflow)
  Downloading Markdown-2.6.8.tar.gz (307kB)
    100% |████████████████████████████████| 317kB 173kB/s 
Requirement already up-to-date: wheel in ./.local/lib/python2.7/site-packages (from tensorflow)
Requirement already up-to-date: six>=1.10.0 in ./.local/lib/python2.7/site-packages (from tensorflow)
Collecting protobuf>=3.2.0 (from tensorflow)
  Downloading protobuf-3.3.0-cp27-cp27mu-manylinux1_x86_64.whl (5.7MB)
    100% |████████████████████████████████| 5.7MB 96kB/s 
Collecting backports.weakref==1.0rc1 (from tensorflow)
  Downloading backports.weakref-1.0rc1-py2-none-any.whl
Requirement already up-to-date: html5lib==0.9999999 in ./.local/lib/python2.7/site-packages (from tensorflow)
Collecting werkzeug>=0.11.10 (from tensorflow)
  Downloading Werkzeug-0.12.2-py2.py3-none-any.whl (312kB)
    100% |████████████████████████████████| 317kB 118kB/s 
Requirement already up-to-date: funcsigs>=1; python_version < "3.3" in ./.local/lib/python2.7/site-packages (from mock>=2.0.0->tensorflow)
Collecting pbr>=0.11 (from mock>=2.0.0->tensorflow)
  Downloading pbr-3.1.1-py2.py3-none-any.whl (99kB)
    100% |████████████████████████████████| 102kB 74kB/s 
Collecting setuptools (from protobuf>=3.2.0->tensorflow)
  Downloading setuptools-36.0.1-py2.py3-none-any.whl (476kB)
    100% |████████████████████████████████| 481kB 122kB/s 
Installing collected packages: numpy, markdown, setuptools, protobuf, backports.weakref, werkzeug, tensorflow, pbr
  Found existing installation: numpy 1.12.0
    Uninstalling numpy-1.12.0:
      Successfully uninstalled numpy-1.12.0
  Running setup.py install for markdown ... done
  Found existing installation: setuptools 34.2.0
    Uninstalling setuptools-34.2.0:
      Successfully uninstalled setuptools-34.2.0
  Found existing installation: protobuf 3.2.0
    Uninstalling protobuf-3.2.0:
      Successfully uninstalled protobuf-3.2.0
  Found existing installation: tensorflow 0.12.1
    Uninstalling tensorflow-0.12.1:
      Successfully uninstalled tensorflow-0.12.1
  Found existing installation: pbr 1.10.0
    Uninstalling pbr-1.10.0:
      Successfully uninstalled pbr-1.10.0
Successfully installed backports.weakref-1.0rc1 markdown-2.6.8 numpy-1.13.0 pbr-3.1.1 protobuf-3.3.0 setuptools-36.0.1 tensorflow-1.2.1 werkzeug-0.12.2

allen@T420:~$ sudo pip install --upgrade tensorflow

[sudo] password for allen:

The directory '/home/allen/.cache/pip/http' or its parent directory is not owned by the current user and the cache has been disabled. Please check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.

The directory '/home/allen/.cache/pip' or its parent directory is not owned by the current user and caching wheels has been disabled. check the permissions and owner of that directory. If executing pip with sudo, you may want sudo's -H flag.

Collecting tensorflow

Downloading tensorflow-1.2.1-cp27-cp27mu-manylinux1_x86_64.whl (34.5MB)

100% |████████████████████████████████| 34.5MB 39kB/s

Requirement already up-to-date: mock>=2.0.0 in ./.local/lib/python2.7/site-packages (from tensorflow)

Collecting numpy>=1.11.0 (from tensorflow)

Downloading numpy-1.13.0-cp27-cp27mu-manylinux1_x86_64.whl (16.6MB)

100% |████████████████████████████████| 16.6MB 60kB/s

Requirement already up-to-date: bleach==1.5.0 in ./.local/lib/python2.7/site-packages (from tensorflow)

Collecting markdown>=2.6.8 (from tensorflow)

Downloading Markdown-2.6.8.tar.gz (307kB)

100% |████████████████████████████████| 317kB 173kB/s

Requirement already up-to-date: wheel in ./.local/lib/python2.7/site-packages (from tensorflow)

Requirement already up-to-date: six>=1.10.0 in ./.local/lib/python2.7/site-packages (from tensorflow)

Collecting protobuf>=3.2.0 (from tensorflow)

Downloading protobuf-3.3.0-cp27-cp27mu-manylinux1_x86_64.whl (5.7MB)

100% |████████████████████████████████| 5.7MB 96kB/s

Collecting backports.weakref==1.0rc1 (from tensorflow)

Downloading backports.weakref-1.0rc1-py2-none-any.whl

Requirement already up-to-date: html5lib==0.9999999 in ./.local/lib/python2.7/site-packages (from tensorflow)

Collecting werkzeug>=0.11.10 (from tensorflow)

Downloading Werkzeug-0.12.2-py2.py3-none-any.whl (312kB)

100% |████████████████████████████████| 317kB 118kB/s

Requirement already up-to-date: funcsigs>=1; python_version < "3.3" in ./.local/lib/python2.7/site-packages (from mock>=2.0.0->tensorflow)

Collecting pbr>=0.11 (from mock>=2.0.0->tensorflow)

Downloading pbr-3.1.1-py2.py3-none-any.whl (99kB)

100% |████████████████████████████████| 102kB 74kB/s

Collecting setuptools (from protobuf>=3.2.0->tensorflow)

Downloading setuptools-36.0.1-py2.py3-none-any.whl (476kB)

100% |████████████████████████████████| 481kB 122kB/s

Installing collected packages: numpy, markdown, setuptools, protobuf, backports.weakref, werkzeug, tensorflow, pbr

Found existing installation: numpy 1.12.0

Uninstalling numpy-1.12.0:

Successfully uninstalled numpy-1.12.0

Running setup.py install for markdown ... done

Found existing installation: setuptools 34.2.0

Uninstalling setuptools-34.2.0:

Successfully uninstalled setuptools-34.2.0

Found existing installation: protobuf 3.2.0

Uninstalling protobuf-3.2.0:

Successfully uninstalled protobuf-3.2.0

Found existing installation: tensorflow 0.12.1

Uninstalling tensorflow-0.12.1:

Successfully uninstalled tensorflow-0.12.1

Found existing installation: pbr 1.10.0

Uninstalling pbr-1.10.0:

Successfully uninstalled pbr-1.10.0

Successfully installed backports.weakref-1.0rc1 markdown-2.6.8 numpy-1.13.0 pbr-3.1.1 protobuf-3.3.0 setuptools-36.0.1 tensorflow-1.2.1 werkzeug-0.12.2

点积，余弦定理，复共轭

Posted on 2017年7月1日, updated on 2017年7月2日 by allenmo

点积：在数学中，数量积（dot product; scalar product，也称为点积）是接受在实数R上的两个向量并返回一个实数值标量的二元运算。它是欧几里得空间的标准内积。

http://baike.baidu.com/link?url=jVfGaAWPAfzatZsQuAztJKPJKJCGbzlMrwiFCNc6i-f4RfQatXL5JGz5IZmGJlQSJu0ReW3y89jmdQYqJkLlA8FxZsqFy-3TtKbNZdf5FngLOl0HM8jClkA5u6_GzKIMyQbVgU5jfLw1QokOGLtJjz9g566dNAhUTBGbV4m4FHi

两个向量a = [a1, a2,…, an]和b = [b1, b2,…, bn]的点积定义为：

a·b=a1b1+a2b2+……+anbn。

使用矩阵乘法并把（纵列）向量当作n×1 矩阵，点积还可以写为：

a·b=a*b^T，这里的b^T指示矩阵b的转置。

设A为 m x p 的矩阵，B为 p x n 的矩阵，那么称 m x n 的矩阵C为矩阵A与B的乘积，记作 C = AB ，其中矩阵中的第 i 行第 j 列元素可以表示为：

例子：

In [1]: import numpy as np

In [2]: a = 2

In [3]: b = 3

In [4]: np.dot(a,b)
Out[4]: 6

In [5]: a = [3, 4]

In [6]: b = [5, 6]

In [7]: np.dot(a,b)
Out[7]: 39

In [8]: a = [[1,2],[3,4]]

In [9]: b = [[1,2],[2,3],[3,4]]

In [10]: np.dot(a,b)
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-10-71281829ef00> in <module>()
----> 1 np.dot(a,b)

ValueError: shapes (2,2) and (3,2) not aligned: 2 (dim 1) != 3 (dim 0)

In [11]: np.dot(b,a)
Out[11]: 
array([[ 7, 10],
 [11, 16],
 [15, 22]])

In [1]: import numpy as np

In [2]: a = 2

In [3]: b = 3

In [4]: np.dot(a,b)

Out[4]: 6

In [5]: a = [3, 4]

In [6]: b = [5, 6]

In [7]: np.dot(a,b)

Out[7]: 39

In [8]: a = [[1,2],[3,4]]

In [9]: b = [[1,2],[2,3],[3,4]]

In [10]: np.dot(a,b)

---------------------------------------------------------------------------

ValueError Traceback (most recent call last)

<ipython-input-10-71281829ef00> in <module>()

----> 1 np.dot(a,b)

ValueError: shapes (2,2) and (3,2) not aligned: 2 (dim 1) != 3 (dim 0)

In [11]: np.dot(b,a)

Out[11]:

array([[ 7, 10],

[11, 16],

[15, 22]])

余弦定理(The Law of Cosines)：

http://baike.baidu.com/item/%E4%BD%99%E5%BC%A6%E5%AE%9A%E7%90%86#2_1

复共轭（complex conjugate):

https://en.wikipedia.org/wiki/Complex_conjugate

Geometric representation of z and its conjugate z̅ in the complex plane. The complex conjugate is found by reflecting z across the real axis.

In mathematics, the complex conjugate of a complex number is the number with equal real part and imaginary part equal in magnitude but opposite in sign.^[1]^[2] For example, the complex conjugate of 3 + 4i is 3 − 4i.

Markdown

Posted on 2017年6月27日 by allenmo

http://www.jianshu.com/p/de9c98bba332

企业年报

Posted on 2017年6月25日 by allenmo

www.szmqs.gov.cn

http://www.szmqs.gov.cn/zxbs/zxbl/

十年

Posted on 2017年6月20日, updated on 2017年6月20日 by allenmo

十年如一天，光阴似箭，一切仿佛发生在昨天，毕业时那份青涩，那份单纯，那份友情依然铭记在心。

这是告别躁动青春的十年，这里走向成熟的十年。

理想依然在心不敢忘记，十年的积累，带着十年的苟且，走过十年的现实路，换个角度看何不是作茧自缚？

生命需要突破，毛毛虫破茧成蝶，小方砖到有冀大鹏，是时间在流过，无声无息，一切看似那么自然，一切却不是理所当然。

十年一轮回，回首邑大这方砖，感觉一直都是的那块方砖，十年前是，十年后今天也是，但愿心态一直是块方砖。

让我记住这一刻吧！天是那么的蓝，云是那么的白，草地依然青绿，心依然，梦常在。感谢这一切，感恩所有人！

win7 快速启动栏恢复

Posted on 2017年5月16日, updated on 2017年5月17日 by allenmo

xp时代的快速启动栏带来方便一直觉得很好用，这下win7里默认不显示了。

已多次搜索解决了这个问题，还是记录一下吧，免得又去查。

右键任务栏，打开”工具栏”->”新建工具栏”。
在打开的对话框的文件夹输入框中输入：%userprofile%\AppData\Roaming\Microsoft\Internet Explorer\Quick Launch
好了，快速启动栏出来了，后面就是调整位置，就不详写了。

labview7.1 the license file is unavailable on localhost

Posted on 2017年5月16日, updated on 2017年5月16日 by allenmo

今天在搞Jabil Test，用labview编写一个dll, 然后用C#去封装成Jabil Test可以调用的dll。

用labview2013, 2016都试过，写出来的dll, C#再封装是成功的，但Jabil Test调用就报错。

后来想想，是不是要用低版本labview试试，于是就安装labview7.1, 但安装完成后一直打不开，报如下错误：

“the license file is unavailable on localhost”

反复折腾下来，才发现是license manager里没有加载license文件，奇怪的是以前安装完就会有的啊，不知道什么情况下搞没了，好吧知道原因为好解决了。

1.在 C:\Program Files (x86)\National Instruments\Shared\License Manager\Bin\nilmUtil.exe路经下打开license manager:

2.把lic文件：C:\Program Files (x86)\National Instruments\Shared\License Manager\Licenses\LabVIEW_LabVIEW_070100.lic加载进去。

labview就可以打开了。

参考了：http://forums.ni.com/t5/forums/v3_1/forumtopicpage/board-id/170/thread-id/454745/page/2

大恒图像 Link

Posted on 2017年5月14日, updated on 2017年5月14日 by allenmo

http://www.daheng-image.com/

今天看Jabil Test里的帮助看到相关的一些函数：

DH_CameraADCControl

DH_CameraAGCControl

DH_CameraBegin

DH_CameraClose

DH_CameraEnableVideoMirror

DH_CameraGetBlankingSize

DH_CameraSetBlanking

DH_CameraSetExposureTime

DH_CameraSetOutput_Window

DH_CameraSetResolution

DH_CameraSetShutterTime

DH_CameraSetSnapMode

DH_CameraSetSnapSpeed

DH_CameraSnapShot

DH_CameraSnapShot_Color

记录一下。

AES SHA 区别

Posted on 2017年5月11日, updated on 2017年5月11日 by allenmo

SHA is a family of “Secure Hash Algorithms” that have been developed by the National Security Agency. There is currently a competition among dozens of options for who will become SHA-3, the new hash algorithm for 2012+.

You use SHA functions to take a large document and compute a “digest” (also called “hash”) of the input. It’s important to realize that this is a one-way process. You can’t take a digest and recover the original document.

AES, the Advanced Encryption Standard is a symmetric block algorithm. This means that it takes 16 byte blocks and encrypts them. It is “symmetric” because the key allows for both encryption and decryption.

https://stackoverflow.com/questions/990705/whats-the-difference-between-sha-and-aes-encryption

http://www.cppblog.com/arthaslee/archive/2010/12/01/135186.html

https://www.quora.com/What-is-the-difference-between-SHA-2-and-AES

http://www.willrey.com/support/ssl_des.html