How to to extrapolate the column values with respect to its counter column values count to a new dataframe?

Set Parameter as index, then call .apply on value and create range from 1 to n+1, then explode and finally reset the index

df.set_index('Parameter')['value'].apply(lambda x: range(1, x+1)).explode().reset_index()

  Parameter value
0     Apple     1
1    Orange     1
2    Orange     2
3     Mango     1
4     Mango     2
5     Mango     3
6     Mango     4
7     Mango     5

You can even avoid setting/resetting index, by assigning column then exploding it:

>>> df.assign(value=df['value'].apply(lambda x: range(1, x+1))).explode('value')
  Parameter value
0     Apple     1
1    Orange     1
1    Orange     2
2     Mango     1
2     Mango     2
2     Mango     3
2     Mango     4
2     Mango     5

You could do this:

new_params = []
new_values = []

for param, value in zip(source_df.parameter, source_df.value):
    new_params.append([param]*value)
    new_values.append(list(range(1,value+1)))

resultant_df = pd.DataFrame({'parameter':new_params, 'value':new_values})

I cannot think of a simpler way using pandas.