Python record (5/19 Update)

Usually OSX has python2 in default setting

Using python3:
https://stringpiggy.hpd.io/mac-osx-python3-dual-install/

==================Hackerank log==================
(0) range()
Notice that when we call range(n), the return is not a list [0,1,2,3,...,n]. It is a range object

  spam = range(10)

  spam[4] = -1

  >> ERROR
  spam = list(range(10))
  spam[4] = -1

  >> OK

(1) Append vs extend??

x = [1, 2, 3]
x.append([4, 5])
print (x)

====>[1, 2, 3, [4, 5]]

x = [1, 2, 3]
x.extend([4, 5])
print (x)  

====>[1, 2, 3, 4, 5]

(1) Any

See:https://www.hackerrank.com/challenges/string-validators/problem

  print 'ab123'.isalnum() # check if all element is alphabet or number
  >>> True
  print 'ab123#'.isalnum()
  >>> False 

  If we wanna check there is ANY element if there is alphabet or number

  c = 'ab123#'

  print(any(c.isalnum() for c in s))

  >>> True

(2) Set:

https://blog.csdn.net/business122/article/details/7541486 

  Copy from a set (if you use b=c, modify on b will affect c)

  we have to copy element by element

  (Of course you can use c = b.copy())

    >>> b   

    set([33, 11, 44, 22])   

    >>> c = [i for i in b]   

    >>> c   

    [33, 11, 44, 22] 

  Method: Add(add a element)/ Update(add multiple element)/ 

  Remove(delete one element)/ Discard(delete one element if exist)/

  intersection/ union/ difference 

A good example comes from here:

https://www.hackerrank.com/challenges/find-second-maximum-number-in-a-list/problem

on line to represent second big element in an unsorted array 

Given list is . The maximum score is , second maximum is  

    print sorted(list(set(nums)))[-2]

https://www.hackerrank.com/challenges/nested-list/forum

python students = [['Harry', 37.21], ['Berry', 37.21], ['Tina', 37.2], ['Akriti', 41], ['Harsh', 39]]
The lowest grade of belongs to Tina. The second lowest grade of belongs to both Harry and Berry,
     for _ in range(int(input())):
        name = input()
        score = float(input())
        table.append([name,score])
    s = set([score for (name, score) in table])
    #print(s)  {41.0, 37.2, 37.21, 39.0}
    s = sorted(list(s))
    #print(s) [37.2, 37.21, 39.0, 41.0]
    second_lowest_score = s[1]
    for (name, score) in sorted(table):
        if score == second_lowest_score:
            print(name) # Berry Harry

(3) map/ filter/ reduce 

https://blog.csdn.net/seetheworld518/article/details/46959871

notice that map in python is totally not the same thing in C++

map in C++ in hash map, map in python means mapping a specific function to a given sequence

Example:

  # Lambda
  print map(lambda x: x % 2, range(7)) #[0,1,2,3,4,5,6]
  >>> [0, 1, 0, 1, 0, 1, 0] 

  # When your function involve multiple sequence

  print map(lambda x , y : x ** y, [2,4,6],[3,2,1]) #2^3 4^2 6^1

  >>> [8, 16, 6]

  # Another  

  print map(lambda x , y : (x ** y, x + y), [2,4,6],[3,2,1])
  >>> [(8, 5), (16, 6), (6, 7)] 

  # If not assign function, it become Zip()

  print map(None, [2,4,6],[3,2,1])
  >>> [(2, 3), (4, 2), (6, 1)]

  # Other example to transfer into list
  map(int, (1,2,3))
  >>> [1, 2, 3]

  # Extract key from diction
  map(int, {1:2,2:3,3:4})
  [1, 2, 3]

  # Apply customize function 

  def u_to_l (s):
    return s.upper()
  print map(u_to_l,'asdfd') 

(4)zip

(5) output format (python3)

  print(summ)

  >> 56.00

  print("{0:.2f}".format(summ))

  >> 56.0

(6)list operation

https://docs.python.org/2/tutorial/datastructures.html

  list.append(x)/  # list.extend(x): add x to end of list

  list.insert(i,x) # insert element x at index i

  list.remove(x)   # erase element of value x

  e = list.pop(i)  # erase element of index i and return

  i = list.index(x)# return the index of element x

  c = list.count(x)# return the number outcome of x in list

  list.sort()      # = sorted(list), in-place

  list.reverse()   # reverse, in-place

(7)String

String in python can not modify the value in-place. See example:

https://www.hackerrank.com/challenges/python-mutations/problem

We can choose (a)convert to list and convert back (b)attach sub-string 

  Example:
    spam = 'I have a pet cat.'
    spam[13] = 'r'
    print(spam)

    >> ERROR

    spam = 'I have a pet cat.'

    li = list(spam)

    li[13] = 'r'

    spam = ''.join(li)
    >>> 'I have a pet rat.'

    spam = 'I have a pet cat.'
    spam = spam[:13] + 'r' + spam[14:]
    print(spam) 

    >>> 'I have a pet rat.'

Split/ Join: separate string to list/ combine string from list 

  li = ['my','name','is','bob']  

  ' '.join(li)  

  >>> 'my name is bob'  

  '..'.join(s)  

  >>> 'my..name..is..bob'  

  b = 'my..name..is..bob' 

  b.split("..")

  >>> ['my', 'name', 'is', 'bob']

  b.split("..",1)  

  >>> ['my', 'name..is..bob'] 

Replace:

  http://www.runoob.com/python/att-string-replace.html 

  see: https://www.hackerrank.com/challenges/capitalize/problem

  If you convert it to list and capitalize and convert back, you will lost space

  Use replace instead

    s = string
    for x in s.split():
        s = s.replace(x, x[0].upper()+x[1:])
    #print(s)
    return s

==================Machine Learning log==================

(histogram() and plot())

install pandas and matplotlib
$pip3 install pandas
$pip3 install matplotlib

use in code:
import pandas
from pandas.plotting import scatter_matrix
import matplotlib.pyplot as plt

list and matrix
http://blog.hhjh.tn.edu.tw/biosomeday/?p=655
    array = [0] * attr_number
    matrix = [array]*attr_number
    print(matrix)
    matrix[1][1] = 1
    > [0, 1, 0, 0]
    [0, 1, 0, 0]
    [0, 1, 0, 0]
    [0, 1, 0, 0]
    use this instead https://www.cnblogs.com/woshare/p/5823303.html
    matrix = [[-1 for i in range(attr_number)] for i in range(attr_number)]
    matrix[1][1] = 1
    > [0, 0, 0, 0]
    [0, 1, 0, 0]
    [0, 0, 0, 0]
    [0, 0, 0, 0]

read file and split:
http://1234.logdown.com/posts/260821-python-split

==================Machine Learning log2==================

1.
Cpp 中常用的 vector.find()，在python中可以：

  3 in [1, 2, 3] # => True

或是

    context = ['1','2','?']
    if ('?' in context): 

        print("got '?'")

2. 需要sorting 一個array，找出前k個在原本array的順序時：

>>> s = [2, 3, 1, 4, 5]
>>> sorted(range(len(s)), key=lambda k: s[k])
[2, 0, 1, 3, 4]
>>> 

   ps. 預設降幂排序，想要升幂： sorted(range(len(s)),reverse=True)

3. 想要把list中的數字同除一個分母：

myList[:] = [x / myInt for x in myList] 

4. 將list 隨機排序：

import random
random.shuffle (lst )

Machine Learning log3

1. 從list當中隨機選擇k個index:
可以重複：
    np.random.choice(150, k)
不重複：
    random.sample(range(150), 10)