pandas-datatypes 1.2.0

A library for getting correct datatype from pandas DataFrame

pandas_datatypes

infer_data_types

A Python function for inferring data types of columns in a pandas DataFrame.

Overview

infer_data_types is a utility function designed to assist in the process of inferring data types of columns within a pandas DataFrame. It offers functionality to handle numeric columns with NaN values by optionally filling them with zeros, and it can automatically convert columns to datetime objects where appropriate.

Usage

import pandas as pd
import pandas_datatypes 

# Create a DataFrame
data = {
    'A': [1, 2, 3],
    'B': ['2022-01-01', '2022-02-01', '2022-03-01'],
    'C': [4.0, None, 6.0],
}

df = pd.DataFrame(data)

# Infer data types
data_types = pandas_datatypes.infer_data_types(df)

print(data_types)

Function Description

def infer_data_types(df: pd.DataFrame, fill_numeric_zeros: bool = True) -> dict:

Infer data types of columns in a DataFrame.

Args:
    df (pd.DataFrame): Input DataFrame.
    fill_numeric_zeros (bool, optional): Whether to fill NaN values in numeric columns with zeros. Defaults to True.

Returns:
    dict: Dictionary containing column names as keys and inferred data types as values.

df: Input DataFrame for which data types are to be inferred.

fill_numeric_zeros: Optional boolean parameter to indicate whether to fill NaN values in numeric columns with zeros. Defaults to True.

Returns a dictionary containing column names as keys and inferred data types as values.