I have a data frame that looks like this:
df = pd.DataFrame({"A":['a','a','a','b','b','c','c','c'],
"B":['he','she',None,'I',None,None,'them','our']})
and I have a list
lst = ['a','b','c']
I would like to filter the dataframe (or get a dictionary) where only the first matching value in column "B" is returned.
my_dict = {'a':'he', 'b':'I','c':'them'}
0 Answer
You can dropna, drop_duplicates, convert to Series and then to_dict:
lst = ['a','b','c']
my_dict = (df.dropna(subset=['B'])
.drop_duplicates(subset=['A'])
.set_index('A')
.loc[lst, 'B'].to_dict()
)
Alternative:
my_dict = df.dropna(subset='B').groupby('A')['B'].first().to_dict()
output: {'a': 'he', 'b': 'I', 'c': 'them'}
这家伙很懒，什么都没留下...