Reading, Writing, and Cleaning Data with Pandas
In this lesson, you'll transform a raw CSV file into a clean, analysis-ready DataFrame — using real data from 76,694 sales exported from PostgreSQL.
🧹 WHAT YOU WILL LEARN
Advanced read_csv — usecols, dtype, nrows, parse_dates (22 MB → 12 MB)
str.strip() — correct 3,757 customer names with extra spaces
rename() — standardize column names
isnull / dropna / fillna — detect and handle missing values
.dt accessor — extract year, month, time from datetime64
drop_duplicates() and outlier detection via IQR
to_csv(index=False) and to_parquet — save the clean DataFrame
📂 DATASET USED
sales.csv
SELECT customer, person_type, seller, sale_date, payment_method, value
FROM public.sales
76,694 records · 2020–2025 · 3,757 customers with extra space · 1,698 outliers
📁 CLASS FILES
aula07_pandas.ipynb
vendas.csv
🔔 If the content was useful, subscribe and activate the bell!
#pandas #python #datascience #datacleaning #dataanalysis