版权声明：王家林大咖2018年新书《SPARK大数据商业实战三部曲》清华大学出版，清华大学出版社官方旗舰店（天猫）https://qhdx.tmall.com/?spm=a220o.1000855.1997427721.d4918089.4b2a2e5dT6bUsM https://blog.csdn.net/duan_zhihua/article/details/83539894
cs 231 Convolutional Networks 卷积网络 Fast layers (Cython extension):
快速卷积和池化层是很有挑战性的，为了使大家不感到痛苦，cs231官网提供了cs231n/fast_layers.py文件中卷积层和池化层前向和后向传播的快速实现。快速卷积实现依赖于Cython扩展。首先要编译它，需要从cs231n目录（D:\PycharmProjects\gitCs231n-camp-fork-2018\assignment\assignment2）运行以下命令：
python setup.py build_ext --inplace运行提示报错： error: Unable to find vcvarsall.bat
running install
running build
running build_ext
.......
error: Unable to find vcvarsall.bat从网上查找相关的资料（https://www.cnblogs.com/yyds/p/7065637.html），解决此问题需安装Visual Studio 2015 （https://blog.csdn.net/FACEFRIENDS/article/details/54096815?utm_source=blogxgwz0）
def get_build_version():
"""Return the version of MSVC that was used to build Python.
For Python 2.3 and up, the version number is included in
sys.version. For earlier versions, assume the compiler is MSVC 6.
"""
prefix = "MSC v."
i = sys.version.find(prefix)
if i == -1:
return 6
i = i + len(prefix)
s, rest = sys.version[i:].split(" ", 1)
majorVersion = int(s[:-2]) - 6
if majorVersion >= 13:
# v13 was skipped and should be v14
majorVersion += 1
minorVersion = int(s[2:3]) / 10.0
# I don't think paths are affected by minor version in version 6
if majorVersion == 6:
minorVersion = 0
if majorVersion >= 6:
return majorVersion + minorVersion
# else we don't know what version of the compiler this is
return Nonepython的版本查询：
D:\vcProjects2018\clion>python
Python 3.6.5 |Anaconda, Inc.| (default, Mar 29 2018, 13:32:41) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.version
'3.6.5 |Anaconda, Inc.| (default, Mar 29 2018, 13:32:41) [MSC v.1900 64 bit (AMD64)]'
>>>从网上下载Visual Studio 2015，安装时候注意要自定义安装，编程语言要选上VC ++，然后开始安装，安装过程时间较长，大约要几个小时，安装以后重启windows。
然后再次执行python setup.py build_ext --inplace，运行成功。
c:\users\lenovo\appdata\roaming\python\python36\site-packages\numpy\core\include\numpy\npy_1_7_deprecated_api.h(12) : Warning Msg: Using deprecated NumPy API, disable it by #defining NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION
.......
im2col_cython.c(30465): warning C4996: '_vsnprintf': This function or variable may be unsafe. Consider using _vsnprintf_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details.
C:\Program Files (x86)\Windows Kits\10\include\10.0.10240.0\ucrt\stdio.h(1396): note: 参见“_vsnprintf”的声明
G:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\BIN\x86_amd64\link.exe /nologo /INCREMENTAL:NO /LTCG /DLL /MANIFEST:EMBED,ID=2 /MANIFESTUAC:NO /LIBPATH:G:\ProgramData\Anaconda3\libs /LIBPATH:G:\ProgramData\Anaconda3\PCbuild\amd64 "/LIBPATH:G:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\LIB\amd64" "/LIBPATH:G:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\ATLMFC\LIB\amd64" "/LIBPATH:C:\Program Files (x86)\Windows Kits\10\lib\10.0.10240.0\ucrt\x64" "/LIBPATH:C:\Program Files (x86)\Windows Kits\NETFXSDK\4.6.1\lib\um\x64" "/LIBPATH:C:\Program Files (x86)\Windows Kits\8.1\lib\winv6.3\um\x64" /EXPORT:PyInit_im2col_cython build\temp.win-amd64-3.6\Release\im2col_cython.obj /OUT:D:\PycharmProjects\gitCs231n-camp-fork-2018\assignment\assignment2\cs231n\im2col_cython.cp36-win_amd64.pyd /IMPLIB:build\temp.win-amd64-3.6\Release\im2col_cython.cp36-win_amd64.lib
im2col_cython.obj : warning LNK4197: 多次指定导出“PyInit_im2col_cython”；使用第一个规范
正在创建库 build\temp.win-amd64-3.6\Release\im2col_cython.cp36-win_amd64.lib 和对象 build\temp.win-amd64-3.6\Release\im2col_cython.cp36-win_amd64.exp
正在生成代码
已完成代码的生成在D:\PycharmProjects\gitCs231n-camp-fork-2018\assignment\assignment2目录下查看自动生成的代码文件im2col_cython.c：
......
#define PY_SSIZE_T_CLEAN
#include "Python.h"
#ifndef Py_PYTHON_H
#error Python headers needed to compile C extensions, please install development version of Python.
#elif PY_VERSION_HEX < 0x02060000 || (0x03000000 <= PY_VERSION_HEX && PY_VERSION_HEX < 0x03030000)
#error Cython requires Python 2.6+ or Python 3.3+.
#else
#define CYTHON_ABI "0_28_2"
#define CYTHON_FUTURE_DIVISION 0
#include <stddef.h>
#ifndef offsetof
#define offsetof(type, member) ( (size_t) & ((type*)0) -> member )
#endif
#if !defined(WIN32) && !defined(MS_WINDOWS)
#ifndef __stdcall
#define __stdcall
#endif
#ifndef __cdecl
#define __cdecl
#endif
#ifndef __fastcall
#define __fastcall
#endif
#endif
#ifndef DL_IMPORT
#define DL_IMPORT(t) t
#endif
#ifndef DL_EXPORT
#define DL_EXPORT(t) t
#endif
#define __PYX_COMMA ,
#ifndef HAVE_LONG_LONG
.......在cs231目录下生成了im2col_cython.cp36-win_amd64文件。
运行快速卷积神经网络：
# -*- coding: utf-8 -*-
# As usual, a bit of setup
import numpy as np
import matplotlib.pyplot as plt
from cs231n.classifiers.cnn import *
from cs231n.data_utils import get_CIFAR10_data
from cs231n.gradient_check import eval_numerical_gradient_array, eval_numerical_gradient
from cs231n.layers import *
from cs231n.fast_layers import *
from cs231n.solver import Solver
#%matplotlib inline
plt.rcParams['figure.figsize'] = (10.0, 8.0) # set default size of plots
plt.rcParams['image.interpolation'] = 'nearest'
plt.rcParams['image.cmap'] = 'gray'
# for auto-reloading external modules
# see http://stackoverflow.com/questions/1907993/autoreload-of-modules-in-ipython
#%load_ext autoreload
#%autoreload 2
def rel_error(x, y):
""" returns relative error """
return np.max(np.abs(x - y) / (np.maximum(1e-8, np.abs(x) + np.abs(y))))
data = get_CIFAR10_data()
for k, v in data.items():
print('%s: ' % k, v.shape)
x_shape = (2, 3, 4, 4)
w_shape = (3, 3, 4, 4)
x = np.linspace(-0.1, 0.5, num=np.prod(x_shape)).reshape(x_shape)
w = np.linspace(-0.2, 0.3, num=np.prod(w_shape)).reshape(w_shape)
b = np.linspace(-0.1, 0.2, num=3)
conv_param = {'stride': 2, 'pad': 1}
out, _ = conv_forward_naive(x, w, b, conv_param)
correct_out = np.array([[[[-0.08759809, -0.10987781],
[-0.18387192, -0.2109216 ]],
[[ 0.21027089, 0.21661097],
[ 0.22847626, 0.23004637]],
[[ 0.50813986, 0.54309974],
[ 0.64082444, 0.67101435]]],
[[[-0.98053589, -1.03143541],
[-1.19128892, -1.24695841]],
[[ 0.69108355, 0.66880383],
[ 0.59480972, 0.56776003]],
[[ 2.36270298, 2.36904306],
[ 2.38090835, 2.38247847]]]])
# Compare your output to ours; difference should be around e-8
print('Testing conv_forward_naive')
print('difference: ', rel_error(out, correct_out))
from scipy.misc import imread, imresize
kitten, puppy = imread('kitten.jpg'), imread('puppy.jpg')
# kitten is wide, and puppy is already square
d = kitten.shape[1] - kitten.shape[0]
kitten_cropped = kitten[:, d//2:-d//2, :]
img_size = 200 # Make this smaller if it runs too slow
x = np.zeros((2, 3, img_size, img_size))
x[0, :, :, :] = imresize(puppy, (img_size, img_size)).transpose((2, 0, 1))
x[1, :, :, :] = imresize(kitten_cropped, (img_size, img_size)).transpose((2, 0, 1))
# Set up a convolutional weights holding 2 filters, each 3x3
w = np.zeros((2, 3, 3, 3))
# The first filter converts the image to grayscale.
# Set up the red, green, and blue channels of the filter.
w[0, 0, :, :] = [[0, 0, 0], [0, 0.3, 0], [0, 0, 0]]
w[0, 1, :, :] = [[0, 0, 0], [0, 0.6, 0], [0, 0, 0]]
w[0, 2, :, :] = [[0, 0, 0], [0, 0.1, 0], [0, 0, 0]]
# Second filter detects horizontal edges in the blue channel.
w[1, 2, :, :] = [[1, 2, 1], [0, 0, 0], [-1, -2, -1]]
# Vector of biases. We don't need any bias for the grayscale
# filter, but for the edge detection filter we want to add 128
# to each output so that nothing is negative.
b = np.array([0, 128])
# Compute the result of convolving each input in x with each filter in w,
# offsetting by b, and storing the results in out.
out, _ = conv_forward_naive(x, w, b, {'stride': 1, 'pad': 1})
def imshow_noax(img, normalize=True):
""" Tiny helper to show images as uint8 and remove axis labels """
if normalize:
img_max, img_min = np.max(img), np.min(img)
img = 255.0 * (img - img_min) / (img_max - img_min)
plt.imshow(img.astype('uint8'))
plt.gca().axis('off')
# Show the original images and the results of the conv operation
plt.subplot(2, 3, 1)
imshow_noax(puppy, normalize=False)
plt.title('Original image')
plt.subplot(2, 3, 2)
imshow_noax(out[0, 0])
plt.title('Grayscale')
plt.subplot(2, 3, 3)
imshow_noax(out[0, 1])
plt.title('Edges')
plt.subplot(2, 3, 4)
imshow_noax(kitten_cropped, normalize=False)
plt.subplot(2, 3, 5)
imshow_noax(out[1, 0])
plt.subplot(2, 3, 6)
imshow_noax(out[1, 1])
plt.show()
#Convolution: Naive backward pass
np.random.seed(231)
x = np.random.randn(4, 3, 5, 5)
w = np.random.randn(2, 3, 3, 3)
b = np.random.randn(2,)
dout = np.random.randn(4, 2, 5, 5)
conv_param = {'stride': 1, 'pad': 1}
dx_num = eval_numerical_gradient_array(lambda x: conv_forward_naive(x, w, b, conv_param)[0], x, dout)
dw_num = eval_numerical_gradient_array(lambda w: conv_forward_naive(x, w, b, conv_param)[0], w, dout)
db_num = eval_numerical_gradient_array(lambda b: conv_forward_naive(x, w, b, conv_param)[0], b, dout)
out, cache = conv_forward_naive(x, w, b, conv_param)
dx, dw, db = conv_backward_naive(dout, cache)
# Your errors should be around e-8 or less.
print('Testing conv_backward_naive function')
print('dx error: ', rel_error(dx, dx_num))
print('dw error: ', rel_error(dw, dw_num))
print('db error: ', rel_error(db, db_num))
#Max-Pooling: Naive forward
x_shape = (2, 3, 4, 4)
x = np.linspace(-0.3, 0.4, num=np.prod(x_shape)).reshape(x_shape)
pool_param = {'pool_width': 2, 'pool_height': 2, 'stride': 2}
out, _ = max_pool_forward_naive(x, pool_param)
correct_out = np.array([[[[-0.26315789, -0.24842105],
[-0.20421053, -0.18947368]],
[[-0.14526316, -0.13052632],
[-0.08631579, -0.07157895]],
[[-0.02736842, -0.01263158],
[ 0.03157895, 0.04631579]]],
[[[ 0.09052632, 0.10526316],
[ 0.14947368, 0.16421053]],
[[ 0.20842105, 0.22315789],
[ 0.26736842, 0.28210526]],
[[ 0.32631579, 0.34105263],
[ 0.38526316, 0.4 ]]]])
# Compare your output with ours. Difference should be on the order of e-8.
print('Testing max_pool_forward_naive function:')
print('difference: ', rel_error(out, correct_out))
#Max-Pooling: Naive backward
np.random.seed(231)
x = np.random.randn(3, 2, 8, 8)
dout = np.random.randn(3, 2, 4, 4)
pool_param = {'pool_height': 2, 'pool_width': 2, 'stride': 2}
dx_num = eval_numerical_gradient_array(lambda x: max_pool_forward_naive(x, pool_param)[0], x, dout)
out, cache = max_pool_forward_naive(x, pool_param)
dx = max_pool_backward_naive(dout, cache)
# Your error should be on the order of e-12
print('Testing max_pool_backward_naive function:')
print('dx error: ', rel_error(dx, dx_num))
#Fast layers
# Rel errors should be around e-9 or less
from cs231n.fast_layers import conv_forward_fast, conv_backward_fast
from time import time
np.random.seed(231)
x = np.random.randn(100, 3, 31, 31)
w = np.random.randn(25, 3, 3, 3)
b = np.random.randn(25,)
dout = np.random.randn(100, 25, 16, 16)
conv_param = {'stride': 2, 'pad': 1}
t0 = time()
out_naive, cache_naive = conv_forward_naive(x, w, b, conv_param)
t1 = time()
out_fast, cache_fast = conv_forward_fast(x, w, b, conv_param)
t2 = time()
print('Testing conv_forward_fast:')
print('Naive: %fs' % (t1 - t0))
print('Fast: %fs' % (t2 - t1))
print('Speedup: %fx' % ((t1 - t0) / (t2 - t1)))
print('Difference: ', rel_error(out_naive, out_fast))
t0 = time()
dx_naive, dw_naive, db_naive = conv_backward_naive(dout, cache_naive)
t1 = time()
dx_fast, dw_fast, db_fast = conv_backward_fast(dout, cache_fast)
t2 = time()
print('\nTesting conv_backward_fast:')
print('Naive: %fs' % (t1 - t0))
print('Fast: %fs' % (t2 - t1))
print('Speedup: %fx' % ((t1 - t0) / (t2 - t1)))
print('dx difference: ', rel_error(dx_naive, dx_fast))
print('dw difference: ', rel_error(dw_naive, dw_fast))
print('db difference: ', rel_error(db_naive, db_fast))
运行结果如下：
runfile('D:/PycharmProjects/gitCs231n-camp-fork-2018/assignment/assignment2/ConvolutionalNetworks.py', wdir='D:/PycharmProjects/gitCs231n-camp-fork-2018/assignment/assignment2')
X_train: (49000, 3, 32, 32)
y_train: (49000,)
X_val: (1000, 3, 32, 32)
y_val: (1000,)
X_test: (1000, 3, 32, 32)
y_test: (1000,)
Testing conv_forward_naive
difference: 2.2121476417505994e-08
D:/PycharmProjects/gitCs231n-camp-fork-2018/assignment/assignment2/ConvolutionalNetworks.py:61: DeprecationWarning: `imread` is deprecated!
`imread` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.
Use ``imageio.imread`` instead.
kitten, puppy = imread('kitten.jpg'), imread('puppy.jpg')
D:/PycharmProjects/gitCs231n-camp-fork-2018/assignment/assignment2/ConvolutionalNetworks.py:68: DeprecationWarning: `imresize` is deprecated!
`imresize` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.
Use ``skimage.transform.resize`` instead.
x[0, :, :, :] = imresize(puppy, (img_size, img_size)).transpose((2, 0, 1))
D:/PycharmProjects/gitCs231n-camp-fork-2018/assignment/assignment2/ConvolutionalNetworks.py:69: DeprecationWarning: `imresize` is deprecated!
`imresize` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.
Use ``skimage.transform.resize`` instead.
x[1, :, :, :] = imresize(kitten_cropped, (img_size, img_size)).transpose((2, 0, 1))

Testing conv_backward_naive function
dx error: 1.1597670719290016e-08
dw error: 2.2471281788358077e-10
db error: 3.3726894077333484e-11
Testing max_pool_forward_naive function:
difference: 4.1666665157267834e-08
Testing max_pool_backward_naive function:
dx error: 3.27562514223145e-12
Testing conv_forward_fast:
Naive: 5.901659s
Fast: 0.032086s
Speedup: 183.933117x
Difference: 4.926407851494105e-11
Testing conv_backward_fast:
Naive: 11.406309s
Fast: 0.074288s
Speedup: 153.541473x
dx difference: 1.949764775345631e-11
dw difference: 8.315279915756832e-13
db difference: 5.47057556442235e-14