python - How can I assign values to a dataframe with a loop if empty series come up

Question

I am filtering a lot in a bigger Dataframe and putting those results in a new column (different values for each row). Sometimes the filtering end up in an empty Series and I get a "ValueError: Incompatible indexer with Series."

Is there a way that in these cases a "Nan" or a "0" is used?

Here is an example of my problem:

  df=pd.DataFrame(np.arange(0,20).reshape(4,5),columns=["A","B","C","D","E"])
    
    for idx in df.index:
         df.loc[idx,"E"]=df.loc[(df.A>10)&(df.B<11)].C.values

Expected result here: Nan or 0

question from:https://stackoverflow.com/questions/65923307/how-can-i-assign-values-to-a-dataframe-with-a-loop-if-empty-series-come-up

Answer 1

There's probably a slicker way to put this together with np.where() or apply(lambda x: x....) but really you just need to test if the returned value is is an empty dataframe/value, and if it is set your column to zero.

for idx in df.index:
    if len(df.loc[(df.A>10)&(df.B<11)].C.values)==0:
        df.loc[idx,"E"] = 0
    else:
        df.loc[idx,"E"]=df.loc[(df.A>10)&(df.B<11)].C.values