sunsided

// A header file to get you set going with Intel SIMD instrinsic programming. 
// All necessary header files are inlucded for SSE2, SSE41, and AVX2
// Macros make the intrinsics easier to read and  generic so you can compile to 
// SSE2 or AVX2 with the flip of a #define


#define SSE2  //indicates we want SSE2
#define SSE41 //indicates we want SSE4.1 instructions (floor and blend is available)
#define AVX2 //indicates we want AVX2 instructions (double speed!) 

sunsided

# Typical setup to include TensorFlow.
import tensorflow as tf

# Make a queue of file names including all the JPEG images files in the relative
# image directory.
filename_queue = tf.train.string_input_producer(
    tf.train.match_filenames_once("./images/*.jpg"))

# Read an entire image file which is required since they're JPEGs, if the images
# are too large they could be split in advance to smaller files or use the Fixed

sunsided

Keybase proof

I hereby claim:

• I am sunsided on github.
• I am sunside (https://keybase.io/sunside) on keybase.
• I have a public key whose fingerprint is 808E 4A07 72DF 6180 5D0F F445 5A01 1B8A 8E78 85DC

To claim this, I am signing this object:

sunsided

// From http://stackoverflow.com/questions/9128519/reading-many-values-from-numpy-c-api
/*
Make

g++ -o matrix_multiply.o -c matrix_multiply.cpp  -I{python include file  for numpy/noprefix.h} -I{python include}  -fPIC -O2 -Wno-deprecated -Wno-write-strings
g++ -o matrix_multiply.so -shared matrix_multiply.o -L{boost lib path} -lz -lm -ldl -lpthread -boost_python

Python code:
import numpy
import matrix_multiply

sunsided

""" Deep Auto-Encoder implementation
	
	An auto-encoder works as follows:

	Data of dimension k is reduced to a lower dimension j using a matrix multiplication:
	softmax(W*x + b)  = x'
	
	where W is matrix from R^k --> R^j

	A reconstruction matrix W' maps back from R^j --> R^k

sunsided

import numpy as np
import tensorflow as tf

# N, size of matrix. R, rank of data
N = 100
R = 5

# generate data
W_true = np.random.randn(N,R)
C_true = np.random.randn(R,N)

sunsided

import numpy as np
import tensorflow as tf

# N, size of matrix. R, rank of data
N = 100
R = 5

# generate data
W_true = np.random.randn(N,R)
C_true = np.random.randn(R,N)

sunsided

// memdjpeg - A super simple example of how to decode a jpeg in memory
// Kenneth Finnegan, 2012
// blog.thelifeofkenneth.com
//
// After installing jpeglib, compile with:
// cc memdjpeg.c -ljpeg -o memdjpeg
//
// Run with:
// ./memdjpeg filename.jpg
//

sunsided

sunsided

def execfile(filename, globals=None, locals=None):
    # http://stackoverflow.com/a/6357418/195651
    exec(compile(open(filename, "rb").read(), filename, 'exec'), globals, locals)