使用pandas melt函数报错‘unhashable type: Index’求解决方案

Hey there, let's break down why you're hitting this error and how to fix it quickly.

The Root Cause #

Your code is passing pandas Index objects directly to the id_vars/value_vars parameters of melt(), and pandas doesn't handle that well. Specifically:

• df_cns.columns[1:] returns an Index object (not a regular list of column names)
• While df_cns.columns[0] is a single column name string, mixing single values with Index objects can trigger unexpected type handling issues that lead to the "unhashable type: Index" error.

The Fix #

Convert those Index references to regular Python lists, which melt() expects. Here are two clean ways to do this:

Option 1: Convert Index to list in-line

df_cns_depivot = pd.melt(
    df_cns,
    id_vars=[df_cns.columns[0]],  # Wrap single column name in a list
    value_vars=df_cns.columns[1:].tolist(),  # Convert Index to list with .tolist()
    value_name="Units"
)

Option 2: Pre-convert all columns to a list first (more readable)

# First turn all column names into a regular list
all_cols = df_cns.columns.tolist()

# Now slice the list for melt() parameters
df_cns_depivot = pd.melt(
    df_cns,
    id_vars=all_cols[0],
    value_vars=all_cols[1:],
    value_name="Units"
)

Why This Works #

Pandas' melt() function is designed to accept column names as strings, lists of strings, or other hashable iterables. By converting the Index object to a plain list, you're giving melt() exactly the input type it expects, eliminating the unhashable type error.

Wrapping the single id_vars column in a list is also a good practice—it makes your code more consistent and avoids edge cases where pandas might misinterpret a single column name as a non-iterable.

内容的提问来源于stack exchange，提问作者bhavesh