20 Essential Pandas Snippets Every Data Analyst Should Know

Summary

Pandas, built on NumPy, is a powerful Python library for data analysis. This article compiles 20 frequently used Pandas code snippets, organized into three sections: basic data information, basic data processing, and DataFrame operations.

Basic Data Information

Read/write CSV and Excel files

pd.read_csv("csv_file")
pd.DataFrame.from_csv("csv_file")
df.to_csv("data.csv", sep=",", index=False)
pd.read_excel("excel_file")
df.to_excel("data.xlsx", sheet_name='a')

Dataset overview df.info() Statistical summary df.describe() Tabular display with tabulate

from tabulate import tabulate
print(tabulate(print_table, headers=headers))

List columns df.columns First/last n rows

df.head(n)
df.tail(n)

Label‑ and position‑based indexing

df.loc[feature_name]
df.loc[[0], ['size']]
df.iloc[n]

Basic Data Processing

Drop missing values df.dropna(axis=0, how='any') Replace missing values df.replace(to_replace=None, value=None) Detect missing values pd.isnull(object) Delete a column df.drop('feature_variable_name', axis=1) Convert to numeric

pd.to_numeric(df["feature_name"], errors='coerce')

Convert to NumPy array

df.as_matrix()

DataFrame Operations

Apply a function to a column

df["height"].apply(lambda h: 2*h)

def multiply(x):
    return x*2
df["height"].apply(multiply)

Rename a column

df.rename(columns={df.columns[2]:'size'}, inplace=True)

Unique values of a column df["name"].unique() Select multiple columns new_df = df[["name", "size"]] Various statistics

df.min()
df.max()
df.idxmin()
df.idxmax()
df.mean()
df.median()
df.corr()
df["size"].median

Sort values df.sort_values(ascending=False) Boolean indexing

df[df["size"] == 5]