assigning column values to empty dataframe with loc

[paolini]

I'm trying to add a column to an existing dataframe. The dataframe can be empty (0 rows) but I want the column to be added anyway.

import pandas as pd
import numpy as np

df = pd.DataFrame(index=np.arange(4))
df.loc[:,'col'] = 42 # this works fine!

df = pd.DataFrame(index=np.arange(0)) # empty dataframe
df.loc[:,'col'] = 42 # ERROR: ValueError: cannot set a frame with no defined index and a scalar
df['col'] = 42 # this works fine but seems deprecated

pd.__version__ is '0.15.2'

[jreback]

By setting the index to a 0-length array, you are actually setting the index.

This is de-facto the same thing as

df = DataFrame()
df.loc[:,'col']

This is simply not allowed setting as you don't have an index length; you can do it in the constructor directly if you really want.

In [9]: df = DataFrame(index=range(5))

In [10]: df
Out[10]: 
Empty DataFrame
Columns: []
Index: [0, 1, 2, 3, 4]

In [11]: df.loc[:,'foo'] = 42

In [12]: df
Out[12]: 
   foo
0   42
1   42
2   42
3   42
4   42

These semantics are not usually what you would do anyhow; use .append

eg.

In [18]: DataFrame().append(Series(42,index=[1,2],name='col'))
Out[18]: 
      1   2
col  42  42

[paolini]

How do I add a column to a possibly empty, already existing dataframe?

[jreback]

In [1]: df = DataFrame()

In [2]: df['foo'] = 1

In [3]: df
Out[3]: 
Empty DataFrame
Columns: [foo]
Index: []

[paolini]

I thought that df['foo'] = 1 was deprecated. Because in some conditions which I'm not able to reproduce in a test, I got this warning: "A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead"

[jreback]

@paolini certainly not. You are getting the warning because you are making changes to something else that is a view. You should check your code. See the documentation on that section. If it is still showing an error, then post your code.

[paolini]

Ok, thanks for your support!

[AwasthiMaddy]

@jreback

In [1]: df = DataFrame()

In [2]: df['foo'] = 1

In [3]: df
Out[3]: 
Empty DataFrame
Columns: [foo]
Index: []

Here Shouldn't the df have value 1 in the column foo.
That is what one wants when he does df['foo'] = 1 instead it is adding the column but not inserting the value.

[Shreya12sahay]

frequent_itemsets = apriori(data,min_support=0.6)
frequent_itemsets
When i am running this in python 3.6 , I am getting this as output:
Empty DataFrame
Columns: [support, itemsets]
Index: []

My code is displaying correct output till this:

import pandas as pd
from mlxtend.preprocessing import TransactionEncoder
from mlxtend.frequent_patterns import apriori
df=pd.read_csv("E:/gro.csv")
print(df)
citrus fruit semi-finished bread margarine margarine.1 ... Unnamed: 28 Unnamed: 29 Unnamed: 30 Unnamed: 31
0 tropical fruit yogurt coffee NaN ... NaN NaN NaN NaN
1 whole milk NaN NaN NaN ... NaN NaN NaN NaN
2 pip fruit yogurt cream cheese meat spreads ... NaN NaN NaN NaN
3 other vegetables whole milk condensed milk long life bakery product ... NaN NaN NaN NaN
4 whole milk butter yogurt rice ... NaN NaN NaN NaN
5 rolls/buns NaN NaN NaN ... NaN NaN NaN NaN
6 other vegetables UHT-milk rolls/buns bottled beer ... NaN NaN NaN NaN
7 pot plants NaN NaN NaN ... NaN NaN NaN NaN
8 whole milk cereals NaN NaN ... NaN NaN NaN NaN
9 tropical fruit other vegetables white bread bottled water ... NaN

te = TransactionEncoder()
te_ary = te.fit(df).transform(df)
data = pd.DataFrame(te_ary, columns=te.columns_)
print(data)
- . 0 1 2 3 4 5 6 7 ... e f g h i m n r s t u
0 True False False False False False False False False False False ... False True False False True False False True True True True
1 True True False False False False False False False False False ... True True False True True True True True True False False
2 False False False False False False False False False False False ... True False True False True True True True False False False
3 False False True False True False False False False False False ... True False True False True True True True False False False
4 True False False False False False False True False False False ... True False False False False True True False False False False
5 True False False False False False False False True False False ... True False False False False True True False False False False
6 True False False False False False False False False True False ... True False False False False True True False False False False
7 True False False False False False False False False False True ... True False False False False True True False False False False
8 True False False False False False False False False False False ... True False False False False True True False False False False
9 True False False False False False False False False False False ... True False False False False True True False False False False

but not getting my output when running this:

frequent_itemsets = apriori(data, min_support=0.6, use_colnames=True)
frequent_itemsets

The code above is not displaying the dataset, rather it is displaying empty dataframe but dataframe is not empty.
Empty DataFrame
Columns: [support, itemsets]
Index: []

Please help me

[rchauhan1]

I want to add N/A value in empty columns, please help