What is the Kernel trick?

Kernel trick arises from speeding up the SVM learning. In the dual version of the optimization function for SVM:

偷得浮生半日闲

竟然有时间读了下 光荣与梦想， 不错。讲述美国从上世纪三四十年代开始的逆袭故事。

Basic Plotting One -- y = sqrt(x) * 10

After grading the homework, I realized that a curve on the scores is a must...

The curve function is designed to be y = sqrt(x) * 10, 0 <= x <= 100. What does it look like? 

1) Google is always the best friend: Type y = sqrt(x) * 10 in the search box, you will see the plot.

2) An alternative is to use R plot function and save the plot in eps:

curve_func <- function(x) { return 10 * sqrt(x) }
setEPS()
postscript('curve.eps')
plot(curve_func, 0, 100, main = "curve_func(x) = 10 * sqrt(x)", xlab = 'original score', ylab = 'curved score')
dev.off()

3) Another way of doing it in Python:
import matplotlib.pyplot as plt
x = [0.01 * z for z in range(10000)]
y = [0.1 * sqrt(z) for z in range(10000)]
plt.plot(x, y)
plt.show()

 

python learning map /reduce

a = [1, 2, 3]
b = [4, 5, 6, 7]
c = [8, 9, 1, 2, 3]
L = map(lambda x:len(x), [a, b, c])

# L == [3, 4, 5]
N = reduce(lambda x, y: x+y, L)
# N == 12
# Or, if we want to be fancy and do it in one line
N = reduce(lambda x, y: x+y, map(lambda x:len(x), [a, b, c]))


I am going to implement a generic MapReduce framework using Python multiprocess module. This module will exploit the multi-core environment better.
For data exchanges, I will use the tmpfiles.

Christmas, new hope!

It is Christmas now.
I want to form some new habits in this coming year.

1. sleep early ( set auto-shutdown in both windows and Linux), read books after the computer is closed
2. research and paper reading. form a plan
3. keep positive
4. the accumulation of the confidence and courage

cut and gawk

cut is one of the most useful commands for text processing.

cut -d. -f1 file # print out the first column of the file

sed:

grep inside vim

The vim now has built-in grep command. To see all the results, just type :copen