How to make empty Pandas DataFrame with named columns and then add a row?-CodePudding

Questions:

How to make empty Pandas DataFrame, name 2 columns in it?
How to add a row for previously created DataFrame?

Can somebody help please? My code below doesn't work properly, it returns strange DataFrame.

My code:

data_set = pd.DataFrame(columns=['POST_TEXT', 'TARGET'])

data_set[data_set.shape[0]] = ['Today is a great day!', '1']

Code result:

CodePudding user response：

data_set = pd.DataFrame(columns=['POST_TEXT', 'TARGET'])

# output
Empty DataFrame
Columns: [POST_TEXT, TARGET]
Index: []

# add row
data_set = data_set.append({"POST_TEXT": 5, "TARGET": 10}, ignore_index=True)

# output
  POST_TEXT TARGET
0         5     10

So to append row you have to define dict where key is name of the column and value is the value you want to append.

If you would like to add row and populate only one column:

data_set = data_set.append({"POST_TEXT": 50}, ignore_index=True)

# output
   POST_TEXT  TARGET
0       50.0     NaN

CodePudding user response：

instead of adding the value post-creation, it is possible to add it during creation:

data_set = pd.DataFrame(columns=['POST_TEXT', 'TARGET'], data=[['Today is a great day!', '1']])