Top 15 functions for Analytics in Python #python #rstats #analytics

Here is a list of top ten  fifteen functions for analysis in Python

1. import (imports a particular package library in Python)
2. getcwd (from os library) – get current working directory
3. chdir (from os) -change directory
4. listdir (from os ) -list files in the specified directory

5. read_csv(from pandas) reads in a csv file

6. objectname.info (like proc contents in SAS or str in R , it describes the object called objectname)
7. objectname.columns (like proc contents in SAS or names in R , it describes the object variable names of the object called objectname)
8. objectname.head (like head in R , it prints the first few rows in the object called objectname)
9. objectname.tail (like tail in R , it prints the last few rows in the object called objectname)
10. len (length)

11. objectname.ix[rows] (here if rows is a list of numbers this     will give those rows (or index) for the object called objectname)

12. groupby -group by a categorical variable

13. crosstab -cross tab between two categorical variables

14. describe – data analysis exploratory of numerical variables
15. corr – correlation between numerical variables

import pandas as pd #importing packages
import os as os

os.getcwd() #current working directory

'/home/ajay/Desktop'

os.chdir('/home/ajay/Downloads') #changes the working directory

os.getcwd()

'/home/ajay/Downloads'

a=os.getcwd()
os.listdir(a) #lists all the files in a directory

diamonds=pd.read_csv("diamonds.csv")
#note header =0 means we take the first row as a header (default) else we can specify header=None

diamonds.info()

<class 'pandas.core.frame.dataframe'="">
Int64Index: 53940 entries, 0 to 53939
Data columns (total 10 columns):
carat      53940 non-null float64
cut        53940 non-null object
color      53940 non-null object
clarity    53940 non-null object
depth      53940 non-null float64
table      53940 non-null float64
price      53940 non-null int64
x          53940 non-null float64
y          53940 non-null float64
z          53940 non-null float64
dtypes: float64(6), int64(1), object(3)
memory usage: 3.9+ MB

diamonds.head()

diamonds.tail(10)

diamonds.columns

Index([u'carat', u'cut', u'color', u'clarity', u'depth', u'table', u'price', u'x', u'y', u'z'], dtype='object')

b=len(diamonds) #this is the total population size
print(b)

53940

import numpy as np

rows = np.random.choice(diamonds.index.values, 0.0001*b)
print(rows)
sampled_df = diamonds.ix[rows]

[45653  7503 47794 12017 46125]

sampled_df

diamonds.describe()

cut=diamonds.groupby("cut")

cut.count()

cut.mean()

cut.median()

pd.crosstab(diamonds.cut, diamonds.color)

diamonds.corr()

Author: Ajay Ohri

http://about.me/ajayohri View all posts by Ajay Ohri