MCPcopy Create free account
hub / github.com/dask/dask / reduction

Method reduction

dask/dataframe/dask_expr/_collection.py:2132–2273  ·  view source on GitHub ↗

Generic row-wise reductions. Parameters ---------- chunk : callable Function to operate on each partition. Should return a ``pandas.DataFrame``, ``pandas.Series``, or a scalar. aggregate : callable, optional Function to operate on

(
        self,
        chunk,
        aggregate=None,
        combine=None,
        meta=no_default,
        token=None,
        split_every=None,
        chunk_kwargs=None,
        aggregate_kwargs=None,
        combine_kwargs=None,
        **kwargs,
    )

Source from the content-addressed store, hash-verified

2130 return new_collection(self.expr.cummin(skipna=skipna))
2131
2132 def reduction(
2133 self,
2134 chunk,
2135 aggregate=None,
2136 combine=None,
2137 meta=no_default,
2138 token=None,
2139 split_every=None,
2140 chunk_kwargs=None,
2141 aggregate_kwargs=None,
2142 combine_kwargs=None,
2143 **kwargs,
2144 ):
2145 """Generic row-wise reductions.
2146
2147 Parameters
2148 ----------
2149 chunk : callable
2150 Function to operate on each partition. Should return a
2151 ``pandas.DataFrame``, ``pandas.Series``, or a scalar.
2152 aggregate : callable, optional
2153 Function to operate on the concatenated result of ``chunk``. If not
2154 specified, defaults to ``chunk``. Used to do the final aggregation
2155 in a tree reduction.
2156
2157 The input to ``aggregate`` depends on the output of ``chunk``.
2158 If the output of ``chunk`` is a:
2159
2160 - scalar: Input is a Series, with one row per partition.
2161 - Series: Input is a DataFrame, with one row per partition. Columns
2162 are the rows in the output series.
2163 - DataFrame: Input is a DataFrame, with one row per partition.
2164 Columns are the columns in the output dataframes.
2165
2166 Should return a ``pandas.DataFrame``, ``pandas.Series``, or a
2167 scalar.
2168 combine : callable, optional
2169 Function to operate on intermediate concatenated results of
2170 ``chunk`` in a tree-reduction. If not provided, defaults to
2171 ``aggregate``. The input/output requirements should match that of
2172 ``aggregate`` described above.
2173 $META
2174 token : str, optional
2175 The name to use for the output keys.
2176 split_every : int, optional
2177 Group partitions into groups of this size while performing a
2178 tree-reduction. If set to False, no tree-reduction will be used,
2179 and all intermediates will be concatenated and passed to
2180 ``aggregate``. Default is 8.
2181 chunk_kwargs : dict, optional
2182 Keyword arguments to pass on to ``chunk`` only.
2183 aggregate_kwargs : dict, optional
2184 Keyword arguments to pass on to ``aggregate`` only.
2185 combine_kwargs : dict, optional
2186 Keyword arguments to pass on to ``combine`` only.
2187 kwargs :
2188 All remaining keywords will be passed to ``chunk``, ``combine``,
2189 and ``aggregate``.

Calls 3

new_collectionFunction · 0.90
CustomReductionClass · 0.90
copyMethod · 0.45