python parquet pandas pyspark pyarrow

• mungingdata python writing parquet pandas pyspark koalas
• conda install pandas pyarrow

  import pandas as pd
  from pathlib import Path
  import pyedna
  
  ednadata = "data"
  signalid = "signal"
  rawbook = Path(r"file_path.xlsx")
  
  data = pyedna.eDNA(ednadata=ednadata, signalid=signalid, inbook=rawbook)
  
  df = (
      data.table.transpose()
      .reindex(index=data.table_columns.index)
      .set_index(pd.MultiIndex.from_frame(data.table_columns))
      .transpose()
  )
  
  pqpath = rawbook.with_stem("_".join([rawbook.stem, ednadata])).with_suffix(".parquet")
  df.to_parquet(pqpath)
  df2 = pd.read_parquet(pqpath)