[python] 넘파이예제

print("# 1 #")
data = (11, 9, 6, 14)
n = len(data)
for i in range(n):
    print(data[i])

#
print()
print("# 2 #")
s = sum(data)
print("sum={0}".format(s))

#
print("\n# 3 #")
print("data = ( ", end="")
for i in range(n):
    print(data[i], end=" ")
print(")")

#
print("\n# 4 #")
for i in range(n):
    if (i==0):
         print("data=( %d"%(data[0]), end="")
    else:
        print(", %d"%(data[i]), end="")
print(" )")

#
print("\n# 5 #")
for freq in data:
  s += freq
print("sum={0}".format(s))
print()

#
print("\n# 6 #")
print("sum=%d"%(s))
print("sum=%5d"%(s))
print("sum=%-5d*"%(s))
print("sum=%05d"%(s))
print("sum=%05d %5.2f"%(s, s))
print("sum=%05d %6.2f"%(s, s))

#
print("\n# 7 #")
def show(list):
    n = len(list)
    for v in list:
        print(v, end=" ")
    print()
 
relativeFreq=[]
for freq in data:
    relativeFreq.append(freq / s)
 
show(relativeFreq)

# initialize list
print("\n# 8 #")
lst = [0 for _ in range(10)]
lst = [0]*10
show(lst)

#
a = np.array([])
a = np.resize(a, 10)
a.size '10
a.shape '(10,)

#
a = np.arange(0, 3)
'a=[0,1,2]

# relative frequency
print("\n# 9 #")
relativeFreq = [0]*len(data)
for idx, freq in enumerate(data):
    relativeFreq[idx] = freq / s
show(relativeFreq)

# label
print("\n# 10 #")
label = ['A', 'B', 'C', 'D']
for idx in range(len(data)):
   print("%1c\t%2d\t%1.3f" %(label[idx], data[idx], relativeFreq[idx]))

# numpy
print("\n# 11 #")
import numpy as np
ndata = np.array([11, 9, 6, 14])
s = ndata.sum()
relativeFreq = [ freq / s for freq in ndata]
for idx in range(ndata.size):
   print("%1c\t%2d\t%1.3f\t%2.2f%%" %(label[idx], ndata[idx], relativeFreq[idx], relativeFreq[idx]*100.0))


# * graph
print("\n# 12 #")
HEIGHT = 10
star = np.array([f * HEIGHT for f in relativeFreq])
show(star)
star = star.round()
star = star.astype(np.int32)
show(star)

space = [HEIGHT-s for s in star]
show(space)

for i in range(HEIGHT):
    c = [i < s for s in space]
    print(c)
    c = [' ' if c[idx]==True else '*' for idx in range(len(star))]
    print(c)


#
print("\n# 13 #") 
for i in range(HEIGHT):
    c = [i < s for s in space]
    c = [' ' if c[idx]==True else '*' for idx in range(len(star))]
    print(c)

 
#
print("\n# 14 #") 
for i in range(HEIGHT):
    c = [i < s for s in space]
    c = [' ' if c[idx]==True else '*' for idx in range(len(star))]
    for j in range(len(star)):
        print("%c "%c[j], end="")
    print()
 
 
# remove empty line
print("\n# 15 #") 
for i in range(HEIGHT):
    c = [i < s for s in space]
    c = [' ' if c[idx]==True else '*' for idx in range(len(star))]
    d = all([c[idx]==' ' for idx in range(len(star))])
    if (d==False):
        for j in range(len(star)):
            print("%c "%c[j], end="")
        print()
 
     
 # another example
print("\n# 16 #")
label = ['seoul', 'pusan', 'daegu', 'incheon', 'kwangju', 'daejeon', 'ulsan']
ndata = np.array([5087, 1639, 1180, 1350, 664, 726, 544])
sum = ndata.sum()
rf = [f/sum for f in ndata]
for idx in range(ndata.size):
  print("%10s\t%2d\t%1.3f\t%2.2f%%" %(label[idx], ndata[idx], rf[idx], rf[idx]*100))
print()
H = 20
star = np.array([f * H for f in rf])
star = star.round()
star = star.astype(np.int32)
star = [H-s for s in star]
for i in range(H):
  c = [' ' if i < s else '*' for s in star]
  d = all([c[idx]==' ' for idx in range(ndata.size)])
  if (d == False):
      for j in range(ndata.size):
          print("%c" %c[j], end='')
      print()
   
   
 # pyplot
print("\n# 17 #")
import matplotlib.pyplot as plt
plt.plot(rf)
plt.bar(range(ndata.size), rf)
plt.xticks(range(ndata.size), label)
plt.show()

# data set -> relative freq., cumulative relative freq. table, histogram
# read data from file, data.csv
# pyplot: http://bcho.tistory.com/m/1201 pyplot 사용법
# numpy: http://taewan.kim/post/numpy_cheat_sheet/