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

Method drop_duplicates

dask/dataframe/dask_expr/_collection.py:3150–3175  ·  view source on GitHub ↗
(
        self,
        subset=None,
        split_every=None,
        split_out=True,
        shuffle_method=None,
        ignore_index=False,
        keep="first",
    )

Source from the content-addressed store, hash-verified

3148 inconsistencies="keep=False will raise a ``NotImplementedError``",
3149 )
3150 def drop_duplicates(
3151 self,
3152 subset=None,
3153 split_every=None,
3154 split_out=True,
3155 shuffle_method=None,
3156 ignore_index=False,
3157 keep="first",
3158 ):
3159 shuffle_method = _get_shuffle_preferring_order(shuffle_method)
3160 if keep is False:
3161 raise NotImplementedError("drop_duplicates with keep=False")
3162 # Fail early if subset is not valid, e.g. missing columns
3163 subset = _convert_to_list(subset)
3164 meta_nonempty(self._meta).drop_duplicates(subset=subset, keep=keep)
3165 return new_collection(
3166 DropDuplicates(
3167 self,
3168 subset=subset,
3169 ignore_index=ignore_index,
3170 split_out=split_out,
3171 split_every=split_every,
3172 shuffle_method=shuffle_method,
3173 keep=keep,
3174 )
3175 )
3176
3177 @insert_meta_param_description(pad=12)
3178 def apply(self, function, *args, meta=no_default, axis=0, **kwargs):

Calls 5

_convert_to_listFunction · 0.90
new_collectionFunction · 0.90
DropDuplicatesClass · 0.90
drop_duplicatesMethod · 0.45

Tested by 4