Author: Ruei-Min Lin

Note: The Python version in my environment is 2.7.9

Base Challenge
--------------------------------
usage
	python ecdf.py --school "Port Chester University" input_file1.csv input_file2.csv
	
algorithm
	given scores[] (of students in a specific school) and percentile p,
	the percentile value (output) should be [0].extend(scores)[p/100 * (N+1)]
	i.e. 
	when p = 0 * 100/(N+1) ~ 1 * 100/(N+1) => output = 0
	when p = 1 * 100/(N+1) ~ 2 * 100/(N+1) => output = scores[0]
	...
	when p = (k-1) * 100/(N+1) ~ k * 100/(N+1) => output = scores[k]
	...
	when p = N * 100/(N+1) ~ (N+1) * 100/(N+1) => output = scores[N-1] highest score among students

	
Bonus Challenge #1
--------------------------------
note
	The client MUST be able to access the network since I use some external resources (jquery files).
	client.py and server.py both use port 8000, and it could be changed easily if port 8000 is occupied:
		server.py:8 PORT = 8000 # change 8000 to whatever port you want to use
		client.py:29 localhost:8000 # change 8000 to whatever port you want to use
	
server usage
	python server.py --school "Port Chester University" input_file1.csv input_file2.csv
	
client usage	
	python client.py X Y
	or using any modern web browser to url http://localhost:8000 (the default port could be configured in server.py)

output
	client would print out what it got from server;
	server would log requests in the file request_history.log
	
	
Bonus Challenge #2
--------------------------------
note
	MUST install plenty of dependent libs such as matplotlib, scipy, statsmodels ...etc.
	It might take an hour to download and install.
	In case that your environment not works, I backup my local result into demo.png.
	The simulation input is generated by test.py
	
usage
	python visual.py --school "Port Chester University" input_file1.csv input_file2.csv