bach.DataFrame.loc

property loc​

The .loc accessor offers different methods for label-based selection. The following are valid use-cases:

Examples

In:
df.to_pandas()
Out:
       values
index
a           1
b           2
c           3
d           4

Getting Values

Single-label selection: Returns all rows where the index column is equal to the label. This returns a Bach Series, where all selected columns are stacked as a single index column.

In:
df.loc['a'].to_pandas()
Out:
__stacked_index
values    1
Name: __stacked, dtype: int64

List-label selection: Returns all rows where the index column is equal to any of the labels to be selected. Returns a Bach DataFrame.

In:
df.loc[['a', 'b']].to_pandas()
Out:
       values
index
a           1
b           2

Slicing selection by labels: Returns all rows between the start and stop of the slice (start and stop are inclusive).

In:
df = df.sort_index()  # slicing is supported only when frame is sorted
df.loc['a':'c'].to_pandas()
Out:
       values
index
a           1
b           2
c           3

Slicing by series boolean: Returns all rows where the series boolean is true.

In:
df.loc[df['values'] == 2].to_pandas()
Out:
       values
index
b           2

For each of the previous types of selection, column selection is supported, this can be by just passing a label, list of labels or a slice. For example:

In:
df['extra_col'] = 1
df.loc['a', 'extra_col'].to_pandas()
Out:
__stacked_index
extra_col    1
Name: __stacked, dtype: int64

Setting Values

Set values for an entire row: Will modify all columns where index is matched.

In:
df.loc['a'] = 2
df.to_pandas()
Out:
       values  extra_col
index
a           2          2
b           2          1
c           3          1
d           4          1

Set values for multiple rows and specific columns: Modifies only the passed columns.

In:
df.loc[['b', 'd'], 'values'] = 10
df.to_pandas()
Out:
       values  extra_col
index
a           2          2
b          10          1
c           3          1
d          10          1

Set values for row slice: Modifies all rows included in the slice.

In:
df.loc['a':'c', 'values'] = 3
df.to_pandas()
Out:
        values   extra_col
index
a           3           2
b           3           1
c           3           1
d          10           1
note
  • .loc supports access-only for first-level index.
  • Slicing rows is supported-only when the dataframe is sorted.