User Tools

Site Tools

Trace:
pandas_series

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
pandas_series [2021/09/15 19:57] – rajupandas_series [2024/02/06 05:18] (current) – [return a random element] raju
Line 1: Line 1:
 +===== creating a series =====
 +==== create a series from a list ====
 +<code>
 +>>> a = pd.Series(['sun', 'mon', 'tue'])
 +>>> a
 +0    sun
 +1    mon
 +2    tue
 +dtype: object
 +</code>
 +
 +To assign an index
 +<code>
 +>>> b = pd.Series(['sun', 'mon', 'tue'], index=['s', 'm', 't'])
 +>>> b
 +s    sun
 +m    mon
 +t    tue
 +dtype: object
 +</code>
 +
 +To assign a name to the column
 +<code>
 +>>> c = pd.Series(['sun', 'mon', 'tue'], index=['s', 'm', 't'], name='day')
 +>>> c
 +s    sun
 +m    mon
 +t    tue
 +Name: day, dtype: object
 +</code>
 +
 +To assign a name to the index
 +<code>
 +>>> d = pd.Series(['sun', 'mon', 'tue'], index=['s', 'm', 't'], name='day')
 +>>> d.index.name = 'letter'
 +>>> d
 +letter
 +s    sun
 +m    mon
 +t    tue
 +Name: day, dtype: object
 +</code>
 +
 +Column name is useful when converting the series to dataframe.
 +<code>
 +>>> b.to_frame()
 +     0
 +s  sun
 +m  mon
 +t  tue
 +
 +>>> c.to_frame()
 +   day
 +s  sun
 +m  mon
 +t  tue
 +</code>
 +
 +If the series did not have a name to begin with but we desire to have one while converting to the dataframe
 +<code>
 +>>> b.to_frame(name='days')
 +  days
 +s  sun
 +m  mon
 +t  tue
 +</code>
 +
 +The index name comes in handy while resetting the index
 +<code>
 +>>> c.reset_index()
 +  index  day
 +0     s  sun
 +1     m  mon
 +2     t  tue
 +>>> d.reset_index()
 +  letter  day
 +0      s  sun
 +1      m  mon
 +2      t  tue
 +</code>
 +
 ===== dummy ===== ===== dummy =====
 ==== append element to series ==== ==== append element to series ====
Line 21: Line 102:
 </code> </code>
  
 +==== return a random element ====
 +Use pandas.Series.sample
 +
 +Ref:-
 +  * https://pandas.pydata.org/docs/reference/api/pandas.Series.sample.html
 +
 +==== expand a series ====
 +tags | using reindex, change index
 +
 +Given two series S, I of length n, and an integer N which is >= n, the idea here is to expand S into an N-element vector, E so that E[I[:]] = S[:].
 +
 +For example if S is [3.4, 1.8], I is [3, 5] and N is 10, we want E to be [0, 0, 0, 3.4, 0, 1.8, 0, 0, 0, 0]
 +
 +<code>
 +import pandas as pd
 +import numpy as np
 +
 +def expand_series(S, I, N, id='val'):
 +    E = pd.Series(S.values, index=I, name=id).reindex(np.arange(0, N)).fillna(0)
 +    return E
 +</code>
 +
 +<code>
 +df = pd.DataFrame({'id': [3,5], 'val': [3.4, 1.8]})
 +print(df)
 +</code>
 +
 +<code>
 +   id  val
 +0   3  3.4
 +1   5  1.8
 +</code>
 +
 +<code>
 +unravelled_series = expand_series(df['val'], df['id'], 10)
 +print(unravelled_series)
 +</code>
 +
 +<code>
 +id
 +0    0.0
 +1    0.0
 +2    0.0
 +3    3.4
 +4    0.0
 +5    1.8
 +6    0.0
 +7    0.0
 +8    0.0
 +9    0.0
 +Name: val, dtype: float64
 +</code>
 +
 +Sample code: https://github.com/KamarajuKusumanchi/notebooks/blob/master/pandas/expand%20a%20series.ipynb
 +
 +Ref:
 +
 +  * https://stackoverflow.com/questions/40029071/setting-series-as-index
 +  * https://chrisalbon.com/python/data_wrangling/pandas_dataframe_reindexing/ - contains some examples on using pandas.Series.reindex
 +  * https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Series.reindex.html - API
 +
 +==== Convert series to a dataframe ====
 +
 +Use to_frame(). By default, it will use the series name to set the column name in the dataframe. But you can also assign one while calling the to_frame function.
 +
 +<code>
 +>>> import pandas as pd
 +>>> b = pd.Series(['sun', 'mon', 'tue'], index=['s', 'm', 't'])
 +>>> b
 +s    sun
 +m    mon
 +t    tue
 +dtype: object
 +>>> b.to_frame()
 +     0
 +s  sun
 +m  mon
 +t  tue
 +>>> c = pd.Series(['sun', 'mon', 'tue'], index=['s', 'm', 't'], name='day')
 +>>> c
 +s    sun
 +m    mon
 +t    tue
 +Name: day, dtype: object
 +>>> c.to_frame()
 +   day
 +s  sun
 +m  mon
 +t  tue
 +>>> b.to_frame(name='days')
 +  days
 +s  sun
 +m  mon
 +t  tue
 +</code>
 ===== check if ===== ===== check if =====
 ==== check if a series is empty ==== ==== check if a series is empty ====
Line 65: Line 241:
  
 Used Python 3.9.4 and IPython 7.22.0 Used Python 3.9.4 and IPython 7.22.0
 +
 +tags | check if a series has at least one element
 +
 +==== check if all elements in a series are unique ====
 +Use pandas.Series.is_unique
 +
 +<code>
 +In [1]: 
 +import pandas as pd
 +
 +In [2]: 
 +pd.Series([1, 2, 3]).is_unique
 +Out[2]: 
 +True
 +
 +In [3]: 
 +pd.Series([1, 2, 2]).is_unique
 +Out[3]: 
 +False
 +</code>
 +
 +Missing values are treated as any other value. So if there are multiple NaNs, it will return True. If this is not desired, drop the NaNs first.
 +<code>
 +In [4]: 
 +import numpy as np
 +pd.Series([1, 2, 3, np.nan, np.nan]).is_unique
 +Out[4]: 
 +False
 +
 +In [5]: 
 +pd.Series([1, 2, 3, np.nan, np.nan]).dropna().is_unique
 +Out[5]: 
 +True
 +</code>
 +
 +For completeness
 +<code>
 +In [6]: 
 +pd.Series([1, 2, 2, np.nan, np.nan]).is_unique
 +Out[6]: 
 +False
 +
 +In [7]: 
 +pd.Series([1, 2, 2, np.nan, np.nan]).dropna().is_unique
 +Out[7]: 
 +False
 +</code>
 +
 +Using | pandas 1.5.3, python 3.11.4, ipython 8.12.0
 +
 +Ref:-
 +  * https://pandas.pydata.org/docs/reference/api/pandas.Series.is_unique.html
 +  * https://stackoverflow.com/questions/48838247/how-to-check-every-pandas-series-value-is-unique
 +
pandas_series.1631735874.txt.gz · Last modified: 2021/09/15 19:57 by raju