Problem with creating DataFrame using Dictionary

Problem with creating DataFrame using Dictionary - Printable Version

+- Python Forum (https://python-forum.io)
+-- Forum: Python Coding (https://python-forum.io/forum-7.html)
+--- Forum: Data Science (https://python-forum.io/forum-44.html)
+--- Thread: Problem with creating DataFrame using Dictionary (/thread-19481.html)

Problem with creating DataFrame using Dictionary - ift38375 - Jul-01-2019

Hi,

When i am creating dataframe for desired table then it is working fine as given below

import pandas as pd
diSales= { 2016:{'qtr1':34500,'qtr2':56000,'qtr3':47000,'qtr4':49000},
           2017:{'qtr1':44900,'qtr2':46100,'qtr3':57000,'qtr4':59000},
           2018:{'qtr1':54500,'qtr2':51000,'qtr3':57000,'qtr4':58500},
           2019:{'qtr1':61000}
         }
ks = pd.DataFrame(diSales)
print(ks)

and output for given code is :

Output:      2016   2017   2018     2019
qtr1  34500  44900  54500  61000.0
qtr2  56000  46100  51000      NaN
qtr3  47000  57000  57000      NaN
qtr4  49000  59000  58500      NaN

But when i am trying to change Column to Mon, Tues , Wed, Thrus then it is giving error

import pandas as pd
diSales= { Mon:{'qtr1':34500,'qtr2':56000,'qtr3':47000,'qtr4':49000},
           Tues:{'qtr1':44900,'qtr2':46100,'qtr3':57000,'qtr4':59000},
           Wed:{'qtr1':54500,'qtr2':51000,'qtr3':57000,'qtr4':58500},
           Thrus:{'qtr1':61000}
         }
ks = pd.DataFrame(diSales)
print(ks)

then it is showing error in output as :

Output:Traceback (most recent call last):
  File "C:\Users\Bhavik\Desktop\agg.py", line 2, in <module>
    diSales= { Mon:{'qtr1':34500,'qtr2':56000,'qtr3':47000,'qtr4':49000},
NameError: name 'Mon' is not defined

Plz solve my problem..

RE: Problem with creating DataFrame using Dictionary - ichabod801 - Jul-01-2019

Mon is a name in Python, not a string. It needs something it is associated with, like an integer or a function or a class. The error is saying you haven't made that association yet. 'Mon' (with quotes) is a string value, suitable to be used as a key in a dictionary.

RE: Problem with creating DataFrame using Dictionary - ift38375 - Jul-02-2019

I want to count particular column occurrence in Dataframe with given below Dataframe ?

import pandas as pd
diSales= { 2016:{'qtr1':34500,'qtr2':56000,'qtr3':47000,'qtr4':49000},
           2017:{'qtr1':44900,'qtr2':46100,'qtr3':57000,'qtr4':59000},
           2018:{'qtr1':54500,'qtr2':51000,'qtr3':57000,'qtr4':58500},
           2019:{'qtr1':61000}
         }
ks = pd.DataFrame(diSales)
ptr = ks.2016.count()
print(ptr)

but when i was trying then showing error " invalid syntax ?
is dataframe allow operation only on alphabetical column names instead of numeric column name ?

RE: Problem with creating DataFrame using Dictionary - ichabod801 - Jul-02-2019

When you were trying what? Show use the code you tried, and show us the syntax error.

RE: Problem with creating DataFrame using Dictionary - ift38375 - Jul-02-2019

(Jul-02-2019, 02:22 AM)ichabod801 Wrote: When you were trying what? Show use the code you tried, and show us the syntax error.

Sir,
I already shared code and error comes with line "ks.2016.count()" bcoz it not takes value 2016 as column.

RE: Problem with creating DataFrame using Dictionary - scidam - Jul-02-2019

(Jul-02-2019, 02:46 AM)ift38375 Wrote: I already shared code and error comes with line "ks.2016.count()" bcoz it not takes value 2016 as column.

Pandas allows to get access to columns of a dataframe instance as its attributes, e.g. df.column_name, but you need to be careful when using this approach. It is better to use .loc selector instead, e.g. df.loc[:, 2016].count().

RE: Problem with creating DataFrame using Dictionary - ift38375 - Jul-02-2019

(Jul-02-2019, 03:22 AM)scidam Wrote:
(Jul-02-2019, 02:46 AM)ift38375 Wrote: I already shared code and error comes with line "ks.2016.count()" bcoz it not takes value 2016 as column.
Pandas allows to get access to columns of a dataframe instance as its attributes, e.g. df.column_name, but you need to be careful when using this approach. It is better to use .loc selector instead, e.g. df.loc[:, 2016].count().

when i am doing same thing with given dataframe then no error comes with alphabetical column name:

import pandas as pd
diSales= { 'Mon':{'qtr1':34500,'qtr2':56000,'qtr3':47000,'qtr4':49000},
           'Tues':{'qtr1':44900,'qtr2':46100,'qtr3':57000,'qtr4':59000},
           'Wed':{'qtr1':54500,'qtr2':51000,'qtr3':57000,'qtr4':58500},
           'Thrus':{'qtr1':61000}
         }
df = pd.DataFrame(diSales)
ks = df.Mon.count()
print(ks)

It gives output '4'

But when i am changing column names and taking numeric form column name as 2016,2017,2018,2019 so on. then it gives error of " Invalid syntax" when i am doing
ks = df.2016.count() .
Why ?

RE: Problem with creating DataFrame using Dictionary - scidam - Jul-02-2019

(Jul-02-2019, 07:11 AM)ift38375 Wrote: Why ?

I don't really know how to answer to the question. It has at least short and long answers. The former is simple -- names of Python objects (names of functions, classes, attributes, instance methods, etc) should not start with a digit (you can think here that you are trying to get access to the instance attribute/or method starting with a digit), the latter will require you to dive into Python grammar and programming language implementation.

RE: Problem with creating DataFrame using Dictionary - ThomasL - Jul-02-2019

I highly suggest you reading and learning pandas basics
https://pandas.pydata.org/pandas-docs/version/0.15/indexing.html