Pandas Data Modification, Iteration, and Function Application Techniques

During data analysis and modeling, Pandas is frequently used to modify, add, or delete fields; this guide introduces common operations and their syntax.

Value modification examples include accessing cells with df.iloc[0,0] , updating values, and bulk changes like df[df.Q1 < 60] = 60 .

Replacement can be performed with df.replace(0, 5) or dictionary mappings such as df.replace({'Q1': 0}, 100) .

Filling missing values uses df.fillna(0) , method options like df.fillna(method='ffill') , or column‑specific dictionaries.

Renaming columns and indexes is done via df.rename(columns={'team':'class'}) and df.rename_axis('animal') .

Adding columns can be achieved with direct assignment df['foo'] = 100 , calculations df['total'] = df.select_dtypes(include=['int']).sum(1) , or df.insert(2, 'total', df.sum(1)) . The df.assign() method allows chaining multiple new columns.

Evaluating expressions uses df.eval('total = Q1+Q2+Q3') and can incorporate variables.

Appending rows is demonstrated with df.loc[100] = ['tom', 'A', 88, 88, 88, 88] and df.append(df2) .

Deletion examples include s.pop(3) and df.dropna() with various parameters.

Advanced filtering shows df.where(df > 70) , np.where(df >= 60, '合格', '不及格') , and df.mask(s > 80) .

Iteration methods covered are for i in df.name , df.iterrows() , df.itertuples() , df.items() , and column iteration for column in df .

Function application includes df.pipe() chaining, df.apply(lambda x: x.lower()) , df.applymap() , df.map() , aggregation with df.agg('max') , transformation via df.transform(lambda x: x*2) , and copying with s.copy() .