Python: how to drop all the non numeric values from a pandas column?

Question

I have a dataframe df like the follwoing

    Rooms   BFS
0   3.5     4201
1   1.5     4201
2   NA      4201
3   NA      4201
4   5.5     4201
5   5       4201
6   4.5     4201
7   3       4201
8           4201
9   3       4201

I want to drop all the non numeric values from the column Rooms . All the values in Rooms are now considered as str

Answer 1

Use to_numeric with errors='coerce' and Series.notna for filtering by boolean indexing :

df = df[pd.to_numeric(df['Rooms'],errors='coerce').notna()]
print (df)
  Rooms   BFS
0   3.5  4201
1   1.5  4201
4   5.5  4201
5     5  4201
6   4.5  4201
7     3  4201
9     3  4201

If need numeric in output first assign to same column and then use DataFrame.dropna :

df['Rooms'] = pd.to_numeric(df['Rooms'],errors='coerce')
df = df.dropna(subset=['Rooms'])
print (df)
   Rooms   BFS
0    3.5  4201
1    1.5  4201
4    5.5  4201
5    5.0  4201
6    4.5  4201
7    3.0  4201
9    3.0  4201