如何将Polars Dataframe转为Python列表及单列转Polars Series

Convert Polars DataFrame to Python List #

Polars provides native methods to convert a DataFrame to a Python list, with options for row-wise or column-wise output:

• Row-wise list: Call to_list() directly on the DataFrame. This returns a list of lists, where each inner list represents the values of a single row.

  import polars as pl
  
  df = pl.DataFrame({"a": [1, 2, 3], "b": [4, 5, 6]})
  row_based_list = df.to_list()
  # Result: [[1, 4], [2, 5], [3, 6]]
  

• Column-wise list: To get each column as a separate list, you can either use to_dict() with as_series=False and extract values, or transpose the DataFrame first before using to_list():

  # Using to_dict
  column_dict = df.to_dict(as_series=False)
  column_based_list = list(column_dict.values())
  # Result: [[1, 2, 3], [4, 5, 6]]
  
  # Using transpose
  column_based_list = df.transpose().to_list()
  # Same result as above
  

Convert Single-Column DataFrame to Polars Series #

For a DataFrame with exactly one column, use these Polars-native methods (pandas commands like squeeze() won't work here):

• Use to_series(): This method directly converts the single-column DataFrame into a Series.

  single_col_df = pl.DataFrame({"a": [1, 2, 3]})
  series = single_col_df.to_series()
  # Output: shape: (3,)
  # Series: 'a' [i64]
  # [
  #  1
  #  2
  #  3
  # ]
  

• Select the column directly: Accessing the column by name or index returns a Series immediately.

  # By column name
  series = single_col_df["a"]
  
  # By column index (0 for the first column)
  series = single_col_df.to_series(0)
  

内容的提问来源于stack exchange，提问作者SKJ