python pandas slice columns

methodRanking:Rank ()Axis index with duplicate valuesThe Is_unique () property of the index can tell you if its value is uniqueSummary and calculation of descriptive statisticsSUM ()Mean ()Describe ()Describing and summarizing statistical functionscorrelation coefficients and covarianceThe series and Dataframe methods are computed for the parameter pairs.Unique value, value count, and membershipUnique value: Unique () methodValue count: The Value_counts () method calculates how often each value

Using Python for data analysis (10) pandas basics: processing missing data, pythonpandasIncomplete Data is common in data analysis. Pandas uses the floating-point value NaN to indicate missing data in floating-point and non-floating-point groups. Pandas uses the isnull () and notnull () functions to determine the missi

. Timestamp (' 20140729 '), ' B ': PD. Series (1, Index=list (range (4))),})Print DF2# You can use Dtypes to see the data formats for each rowPrint Df2.dtypes# then look at how to view the data in the data frame and see all the dataPrint DF# Use Head to see the first few rows of data (default is the first 5 rows), but you can specify the first few linesPrint Df.head ()# View the first three rows of dataPrint Df.head (3)# Use Tail to view the following 2 rows of dataPrint Df.tail (2)# View the in

7.0 8 df 0 1 2 0 NaN NaN 2 1 NaN NaN 5 2 6.0 7.0 8 df.fillna({0:111:22}, inplace=True) 0 1 2 0 11.0 22.0 2 1 11.0 22.0 5 2 6.0 7.0 8 df 0 1 2

The processing of the data is pandas, but it has not been learned and does not know whether there is a method call that is directly normalized to a column. Himself dealing things down. The feeling is still more troublesome.After reading to the array using pandas, I want to have the ' monthlyincome ' column normalized, and the chestnuts on the web are normalized to the entire dataframe, because some of my da

Use Python for data analysis _ Pandas _ basic _ 2, _ pandas_2Reindex method of Series reindex In [15]: obj = Series([3,2,5,7,6,9,0,1,4,8],index=['a','b','c','d','e','f','g', ...: 'h','i','j'])In [16]: obj1 = obj.reindex(['a','b','c','d','e','f','g','h','i','j','k'])In [17]: obj1Out[17]:a 3.0b 2.0c 5.0d 7.0e 6.0f 9.0g 0.0h 1.0i 4.0j 8.0k NaNdtype: float64 If the current va

="bfill"))‘‘‘------Back fill------One, threea-0.211055-2.869212 0.022179b-0.870090-0.878423 1.071588c-0.870090-0.878423 1.071588d-0.203259 0.315897 0.495306e-0.203259 0.315897 0.495306f 0.490568-0.968058-0.999899g 1.437819-0.370934-0.482307H 1.437819-0.370934- 0.482307 ‘‘‘Print ('------Average fill------') Print (Df.fillna (Df.mean ()))‘‘‘------Average fill------One, threea-0.211055-2.869212 0.022179b 0.128797-0.954146 0.021373c-0.870090-0.878423 1.071588d 0.128797-0.95

the unique value of A, the number of occurrences (a, b) of the unique value of statistics = (1,3) c appears 1 times (A, B) = (2,4) appears 3 times - the Print(Pd.crosstab (df['A'],df['B'],normalize=true))#display in a frequency-based manner - Print('--------') - Print(Pd.crosstab (df['A'],df['B'],values=df['C'],aggfunc=np.sum))#values: A value array based on a factor aggregation - #Aggfunc: If the values array is not passed, the frequency table is computed, and if the array is passed, the calc

Two data structure series and dataframe.SeriesThe series is the same as a list in Python, with data and index values.Here we create a series object. Data values and indexes for series objects:The index of the list starts at 0, and the series is indexed by default, similar to the list starting with 0. However, you can also customize the index:Indexes can be redefined:Operation elements according to index:Series is also used in the form of dictionaries:

Pandas mainly has 4 of the time-related types. Timestamp, Period, Datetimeindex,periodindex.ImportPandas as PDImportNumPy as NP##TimestampPd. Timestamp ('9/1/2016 10:05am')#output:timestamp (' 2016-09-01 10:05:00 ')##PeriodPd. Period ('1/2016')#output:period (' 2016-01 ', ' M ')Pd. Period ('3/5/2016')#output:period (' 2016-03-05 ', ' D ')##DatetimeindexT1 = PD. Series (List ('ABC'), [PD. Timestamp ('2016-09-01'), PD. Timestamp ('2016-09-02'), PD. Time

Delete one or more columns of Pandas Dataframe:method One : Direct del df[' Column-name ']method Two : Using the Drop method, there are three types of equivalent expressions:1. df= df.drop (' column_name ', 1);2. Df.drop (' column_name ', Axis=1, Inplace=true)3. Df.drop ([df.columns[[0,1, 3]], axis=1,inplace=true) # Note:zero indexedNote : Usually there is a inplace optional parameter that modifies the orig

Using Python for data analysis (13) pandas basics: Data remodeling/axial rotation, pythonpandas Remodeling DefinitionRemodeling refers to re-arranging data, also called axial rotation.DataFrame provides two methods: Stack: rotate the column of data into rows. Unstack: "Rotate" data rows as columns. For example: Process stack formatThe stack format is also

Getting started with Python for data analysis--pandas Based on the NumPy established from pandas importSeries,DataFrame,import pandas as pd One or two kinds of data structure 1. Series A python-like dictionary with indexes and values

Objective Pandas is a numpy built with more advanced data structures and tools than the NumPy core is the Ndarray,pandas is also centered around Series and dataframe two core data structures. Series and Dataframe correspond to one-dimensional sequence and two-dimensional table structure respectively. Pandas's conventional approach to importing is as follows: From

data conversion refers to filtering, cleaning, and other conversion operations on the data. Remove Duplicate data Repeating rows often appear in the Dataframe, Dataframe provides a duplicated () method to detect whether rows are duplicated, and another drop_duplicates () method to discard duplicate rows:Duplicated () and Drop_duplicates () methods defaultJudging all Columns, if you do not want to, the collection of incoming

3.0 2.0# 1 NaN 3.0 5.0# 2 NaN 4.0 5.0Print' Right ')data = Df.fillna (method=' Ffill ', axis=1)Print (data)# 0 1 2# 0 1.0 1.0 2.0# 1 NaN 3.0 3.0# 2 NaN 4.0 5.0Print' Down ')data = Df.fillna (method=' Ffill ')Print (data)# 0 1 2# 0 1.0 NaN 2.0# 1 1.0 3.0 2.0# 2 1.0 4.0 5.0Print' Left ')data = Df.fillna (method=' Backfill ', axis=1)Print (data)# 0 1 2# 0 1.0 2.0 2.0# 1 3.0 3.0 NaN# 2 4.0 4.0 5.0Print' Up ')data = Df.fillna (method=' Backfill ')Print (data)# 0 1 2# 0 1.0 3.0 2.0# 1 NaN 3.0 5.0# 2

Label:Read the contents of the table, as in the following example: ImportMySQLdbTry: Conn= MySQLdb.connect (host='127.0.0.1', user='Root', passwd='Root', db='MyDB', port=3306) DF= Pd.read_sql ('select * from test;', con=conn) Conn.close ()Print "Finish Load DB" exceptmysqldb.error,e:PrintE.ARGS[1] Write the data to the table, as in the following example DF = PD. DataFrame ([[1,'XXX'],[2,'yyy']],columns=list ('AB')) Try: Conn= MySQLdb.connect (host='1

import NumPy as NPImport Pandas as PD1 #string Common methods-strip2s = PD. Series (['Jack','Jill','Jease','Feank'])3DF = PD. DataFrame (Np.random.randn (3,2), columns=['Column A','Column B'],index=range (3))4 Print(s)5 Print(df.columns)6 7 Print('----')8 Print(S.str.lstrip (). Values)#Remove the left space9 Print(S.str.rstrip (). Values)#Remove the space on the rightTenDf.columns =Df.columns.str.strip () O

1.1. Pandas Analysis steps Loading data COUNT the date of the access_time. SQL similar to the following: SELECT date_format (access_time, '%H '), COUNT (*) from log GROUP by Date_format (access_time, '%H '); 1.2. Code Cat pd_ng_log_stat.py#!/usr/bin/env python#-*-Coding:utf-8-*-From Ng_line_parser import NglineparserImport Pandas as PDImport socketImport str

Close 2017-11-24 260.359985 2017-11-27 260.230011 2017-11-28 262.869995"""if __name__=='__main__': Test_run ()There is a simpy-to-drop the data which index is not present in Dspy:Df1=df1.join (Dspy, how='inner')We can also rename the ' Adj Close ' to prevent conflicts: # Rename the column Dspy=dspy.rename (columns={'Adj Close'SPY'})Load More stocks:ImportPandas as PDdefTest_run (): start_date='2017-11-24'End_data='2017-11-28'dates=Pd.date_range