How to Fix the 'AttributeError: 'DataFrame' object has no attribute 'append'' in Pandas?

AttributeError: 'DataFrame' Object Lacks 'Append' Method

When attempting to incorporate a dictionary into a DataFrame, you may encounter the "AttributeError: 'DataFrame' object has no attribute 'append'" error. While DataFrame does possess an "append" method, it was eliminated in pandas 2.0.

Resolution:

To rectify this issue, utilize the "concat" method as a replacement:

df = pd.concat([df, pd.DataFrame([new_row])], ignore_index=True)

Alternatively, if your DataFrame index is a RangeIndex, you may employ the "loc" method:

df.loc[len(df)] = new_row # Only applicable with RangeIndex

Rationale for Removal:

The removal of the "append" method was motivated by its inefficiency in repeated insertion scenarios. Unlike Python's "list.append" function, which operates in-place, pandas' "append" method constructs a new DataFrame. This difference results in a time complexity of O(n) for pandas' "append," rendering it inefficient.

Alternative Approach:

If you need to repeatedly add items, avoid using "append" or "concat" multiple times. Instead, accumulate new rows in a list and convert it to a DataFrame once the loop is complete. Subsequently, concatenate the extended DataFrame to the original. This approach ensures optimal performance.

For example:

lst = []

for new_row in items_generation_logic:
    lst.append(new_row)

# Create extension
df_extended = pd.DataFrame(lst, columns=['A', 'B', 'C'])

# Concatenate to original
out = pd.concat([df, df_extended])

The above is the detailed content of How to Fix the 'AttributeError: 'DataFrame' object has no attribute 'append'' in Pandas?. For more information, please follow other related articles on the PHP Chinese website!