pandas 0.25.x to 1.5.2, nested renamer migration - pandas

Question

[In]:
pd.set_option('display.max_colwidth', 200)
topic_stats_df = corpus_topic_df.groupby('Dominant Topic').agg({
'Dominant Topic': {
'Doc Count': np.size,
'% Total Docs': np.size }
})
topic_stats_df = topic_stats_df['Dominant Topic'].reset_index()
topic_stats_df['% Total Docs'] = topic_stats_df['% Total Docs'].apply(lambda row: round((row*100) / len(papers), 2))
topic_stats_df['Topic Desc'] = [topics_df.iloc[t]['Terms per Topic'] for t in range(len(topic_stats_df))]
topic_stats_df
[Out]:
---------------------------------------------------------------------------
SpecificationError Traceback (most recent call last)
Cell In[47], line 2
1 pd.set_option('display.max_colwidth', 200)
----> 2 topic_stats_df = corpus_topic_df.groupby('Dominant Topic').agg({
3 'Dominant Topic': {
4 'Doc Count': np.size,
5 '% Total Docs': np.size }
6 })
7 topic_stats_df = topic_stats_df['Dominant Topic'].reset_index()
8 topic_stats_df['% Total Docs'] = topic_stats_df['% Total Docs'].apply(lambda row: round((row*100) / len(papers), 2))
File ~/miniconda3/envs/nlp/lib/python3.8/site-packages/pandas/core/groupby/generic.py:894, in DataFrameGroupBy.aggregate(self, func, engine, engine_kwargs, *args, **kwargs)
891 func = maybe_mangle_lambdas(func)
893 op = GroupByApply(self, func, args, kwargs)
--> 894 result = op.agg()
895 if not is_dict_like(func) and result is not None:
896 return result
File ~/miniconda3/envs/nlp/lib/python3.8/site-packages/pandas/core/apply.py:169, in Apply.agg(self)
166 return self.apply_str()
168 if is_dict_like(arg):
--> 169 return self.agg_dict_like()
170 elif is_list_like(arg):
171 # we require a list, but not a 'str'
172 return self.agg_list_like()
File ~/miniconda3/envs/nlp/lib/python3.8/site-packages/pandas/core/apply.py:478, in Apply.agg_dict_like(self)
475 selected_obj = obj._selected_obj
476 selection = obj._selection
--> 478 arg = self.normalize_dictlike_arg("agg", selected_obj, arg)
480 if selected_obj.ndim == 1:
481 # key only used for output
482 colg = obj._gotitem(selection, ndim=1)
File ~/miniconda3/envs/nlp/lib/python3.8/site-packages/pandas/core/apply.py:594, in Apply.normalize_dictlike_arg(self, how, obj, func)
587 # Can't use func.values(); wouldn't work for a Series
588 if (
589 how == "agg"
590 and isinstance(obj, ABCSeries)
591 and any(is_list_like(v) for _, v in func.items())
592 ) or (any(is_dict_like(v) for _, v in func.items())):
593 # GH 15931 - deprecation of renaming keys
--> 594 raise SpecificationError("nested renamer is not supported")
596 if obj.ndim != 1:
597 # Check for missing columns on a frame
598 cols = set(func.keys()) - set(obj.columns)
SpecificationError: nested renamer is not supported
The code is credited to Sarkar, D. (2019). Text Analytics with Python Apress, Topic modeling section.
Pip pandas 0.25.3 fails because I'm on an m1 Mac.
Have tried: pip install pandas==0.25.3
Have tried: arch -x86_64 pip install pandas==0.25.3

Answer 1

Pandas a will back removed nested renaming in favor of using pd.NamedAgg
topic_stats_df = corpus_topic_df.groupby('Dominant Topic').agg({
'Dominant Topic': {
'Doc Count': np.size,
'% Total Docs': np.size }
})
This statement can be rewritten as follows:
topic_stats_df = corpus_topic_df.groupby('Dominant Topic')\
.agg(Doc_count=('Dominant Topic', np.size),
Pct_Total_Docs=('Dominant Topic', np.size))