Summary_Library🍑

🍒🍒🍒PYTHON LIBRARY NUMBY
0.import numpy

• import numpy as np

1. Make:
   arr = np.array(list)
   or arr = np.array([[1,2,3],[4,5,6]])
2. Characteritics

• Can do sum
  sum(arr1+arr2)
• count column and row:(arr(row. column))
  arr.shape
• count which type of array (1,2,or 3)
  arr.ndim
  -count how many indicator inside:
  arr.size
  -to change the datatype:
  arr = np.array(list, dtype = np.int64)
  arr = arr.astype(np.int64)
  -to change the shape:
  arr = arr.reshape(row,column)
  -to range from..to....
  arr = np.arange(startno,tono)
  -to get random no
  np.random.rand(row,column)
  arr= np.random.randint(st,en,size=(row,column))

🍓🍓🍓🍓LIBRARY PANDAS

• import pandas as pd

1. Make:
   ham 1 chieu: pop = pd.Series([1,2,3,4], index = [de muc], name ='')
   or mang 2 chieu tro len:
   df = pd.DataFrame({key:value})
   =pd.DataFrame(data)
2. Characteritics

• Can do sum
  df.sum(axis =1/0)
• count column and row:(arr(row. column))
  df.shape
• count which type of array (1,2,or 3)
  df.ndim
  -count how many indicator inside:
  df.size
  -to change the datatype:
  df = arr.astype(np.int64)
  -to change the shape:
  df = df.reshape(row,column)
  -to range from..to....
  df = pd.arange(startno,tono)
  -to get random no
  np.random.rand(row,column)
  arr= pd.random.randint(st,en,size=(row,column))
• to delete
  del df[' 1']
  df.drop('javascript',inplace = True)
• to sort value:
  df.sort_values(by = [' ',' '])
  df.sort_values(by = (' ')
• to sort index
  df.sort_index(ascending = False/True): 인덱스 기준으로 절렬
• to import data from excel/csv
  score = pd.read_csv("./data/score.csv", encoding ="cp949", index_col = "과목")
• to add into df
  df.[' '] = [ ]
• to display(lon nhat toi nho nhat)
  display(score.sort_values(by ='2반')[::-1])
• to sum in df and add sum as a column
  score['sum']= score.loc[:,:'4반'].sum(axis = 1)
• to cal min/max
  score.loc["Max score"]=score.loc[:,:'4반'].max(axis =1)
  score.loc["Min score"]=score.loc[:,:'4반'].min(axis =1)
• to split df
  df['Domain'] = df['Email'].str.split('@').str[1]
• to use the function
  def assign_grade(score):
  if score>= 80 :
  return 'A'
  elif score >= 60 :
  return 'B'
  elif score >= 40 :
  return 'C'
  elif score >= 20 :
  return 'D'
  else :
  return 'F'
  score.applymap(assign_grade)
• to merge
  pd.merge(h_info,r_info,how = 'left/right/innner',on = "common thing")
• to groupby
  age_df.groupby(by=["category"]).count/sum()
• to concat (merge + arrange order)
  pd.concat([cnt2019.p2019_2020.cnt2020.p2020_2021.cnt2021],axis=1)
• to rename
  cnt2019.name = "2019 total"

🍏🍏🍏PYTHON_LIBRARY_CHART(MATPLOTLIB)
0. IMPORT
import matplotlib.pyplot as plt
import random
import random as rand
df = pd.DataFrame(np.random.rand(50,4)*100, columns = ['A','B','C','D']).astype(np.int64)

1. MAKE
   x = [row]
   y = [column]
   (line)
   plt.plot(x,y, ls = '-.',color = 'green', lw = 1, marker = '$*$',mec = 'green' , mfc = 'green')
   (shape - bar)
   plt.bar(x,y,color = 'lightblue')
   (shape-bar- opposite)
   plt.barh(x,y)
   (scatter)
   plt.scatter(x = df['A'], y = df['B'], label = 'group1',s = 90)
   plt.scatter(x = df['C'], y = df['D'], label = 'group2')

• to make label
  plt.xlabel('')
  plt.ylabel(' ')
• to make limited(range)
  plt.lim(begin number, end number)
• to make title
  plt.title('title name',loc = 'center/left/right')
• to show the chart
  plt.show()
• to show data head,tail
  data.head()
  data.tail()
• to count how many for each element
  data['seach column name'].value_counts()
• to index the row
  x= y.index
• to show text in the chart
  for i in range (len(y)):
  plt.text(x[i],y[i],f'{y[i]}')
• to larger the size
  plt.figure(figsize = (20,10))
• to sort the largest to smallest
  dt = dt.sort_values('사상자수', ascending = False)