# Pandas Cheatsheet # Cheatsheet 1 In **pandas**, there are a few quick + reliable ways to check the **data type of columns**, depending on what you need. --- #### 1️⃣ Check *all* column types (most common)
`df.dtypes`
**Output example**
`age int64height float64name objectactive booldate datetime64[ns]dtype: object`
--- #### 2️⃣ Detailed overview (very useful for debugging)
`df.info()`
Shows: - column name - non-null count - dtype - memory usage --- #### 3️⃣ Check type of a *single column*
`df["age"].dtype`
or
`df.age.dtype`
--- #### 4️⃣ Check only numeric / non-numeric columns
`df.select_dtypes(include="number").columns`
`df.select_dtypes(exclude="number").columns`
--- #### 5️⃣ Check if a column is a specific type
`import pandas as pdpd.api.types.is_numeric_dtype(df["age"])pd.api.types.is_string_dtype(df["name"])pd.api.types.is_datetime64_any_dtype(df["date"])pd.api.types.is_bool_dtype(df["active"])`
--- #### 6️⃣ Count how many columns per dtype
`df.dtypes.value_counts()`
--- #### 7️⃣ Detect mixed or problematic types (object columns)
`df.select_dtypes(include="object").apply( lambda s: s.map(type).value_counts())`
Useful when: - CSV imported badly - numbers stored as strings - mixed types in one column
TaskCommand
See all dtypes`df.dtypes`
Full summary`df.info()`
Single column`df[col].dtype`
Only numeric`select_dtypes()`
Type check`pd.api.types.is_*`