dossier parquet unique df dataframe
import pandas as pd
import glob
files = glob.glob('Dir/**/*.parquet')
df = pd.concat([pd.read_parquet(fp) for fp in files])
Important Impala