# - * - Coding: UTF-8 - * - '' ' Author: When Yadong function: pandas Application Version: Time: 2019-10-01 ' '' # module import Import PANDAS AS pd Import numpy AS NP Import matplotlib. AS PLT pyplot '' ' Serise data types - one-dimensional data ' '' a = pd.Series ([. 1, 0.3 , np.nan]) B = pd.Series (np.array ([. 1, 2,. 3 ] )) Print ( ' A \ n- ' , A) Print ( ' B \ n- ' , B) # modified index A = pd.Series ([. 1, 0.3, np.nan], index = [ 'a' , ' B ' , ' C ' ]) Print (A) ' '' DataFrame data type - two-dimensional data '' ' # DATE_RANGE () randomly generates a time series DATE = pd.date_range ( ' 20,191,001 ' , periods. 5 = ) # print (DATE) # create objects using numpy , index = DATE, Columns = List (DF = pd.DataFrame (np.random.randn (. 5,. 4) ' ABCD ' )) # print (DF) # View data print (df.head ()) # get the first few rows, five rows before returning to the default Print (df.tail ()) # Get after a few rows, five rows to return after default Print (df.index) # Gets the index # Print (List (df.index)) Print (df.columns) # Gets column name Print (df.values) # Get all value Print (df.describe ()) # acquires description information Print (df.T) # transpose Print (df.sort_index (Axis =. 1, Ascending = False)) # of index objects reordering Print (df.sort_values ( = by ' D ' )) # sort the elements of a column for Print ( ' *' * 50 ) # Select Data Print (DF [ ' A ' ]) # get all the data for a column Print (DF [1: 3]) # Gets the index 1: 3 line data Print (DF [ ' 20,191,001 ' : ' 20191004 ' ]) # Gets the index value of '20191001': '20191004' line data Print ( ' * ' * 50 ) # LOC positioning element method Print (df.loc [date [0]]) # acquisition date first index data Print (df.loc [:, [' A ' , ' B ' ]]) # Gets column named A, B of all the line data Print (df.loc [ ' 20,191,002 ' : ' 20,191,004 ' , [ ' A ' , ' B ' ]]) # Gets the index value of '20191002': '20191004' range of a, column B data Print (df.loc [ ' 20191002 ' , [ ' a ' , ' B ' ]]) # Gets the index value of '20191002'A, B, column data Print ( '* ' * 50 ) # get the data Boolean Print (DF [df.A> 0]) # acquired data in column A is larger than 0 Print (DF [DF> 0]) # obtain all the data is greater than 0 # assignment # Print (DF) S1 = pd.Series ([1,2,3,4], index = pd.date_range ( ' 20,191,002 ' , = periods. 4)) # generates a data type Series # Print ( 'S1 \ n-', s1) DF [ ' F. ' ] = s1 # added back to s1 DF # RINT ( 'DF \ n-', DF) df.at [DATE [0], ' A ' ] = 0 #Alternatively data in the specified table # Print ( 'DF \ n-', DF) df.loc [:, ' D ' ([. 5] * len (DF))] = np.array # specified value in a column to be replaced , array type # Print ( 'DF \ n-', DF) # processing NaN worth embodiment Print (df.dropna (= How ' the any ' )) # deletes all of the data rows containing NaN Print (df.fillna (value =. 3) ) # default value filled NaN3 Print (pd.isnull (DF)) # determines whether or not containing NaN, returns a Boolean value