Personal addition to pandas package for faster and better performance of data wrangling.
Main functions:
1) pandas profile (from mb_pandas.src.profiler import create_profile)
2) df load using async for larger datasets (from mb_pandas.src.dfload import load_any_df)
3) df compare (from mb_pandas.src.profiler import profile_compare)
4) df transformation for basic function (from mb_pandas.src.tranform import *)
['check_null','remove_unnamed','rename_columns','check_drop_duplicates','get_dftype'])
Scripts:
1) df_profile - to create profile for any df. (default folder : /home/malav/pandas_profiles/)
2) df_view - to view the csv or parquet file
Pip install :
pip install mb_pandas
Load all the other packages with mb_base
pip install mb_base
import mb.pandas as pd