📎 This article includes 3 downloadable practice files ↓

📘 Python Beginner Course · Lesson 8 of 12

pandas is the reason so many analysts learn Python. A DataFrame is a table with named columns, like an Excel Table, and operations on whole columns take one line.

Read data

import pandas as pd
df = pd.read_csv("sales.csv", parse_dates=["date"])
# Excel: pd.read_excel("file.xlsx", sheet_name="Data")   (needs openpyxl)

Look before you leap

df.shape          # (120, 7)  rows, columns
df.dtypes         # the type of each column
df.head(3)        # first 3 rows   (df.tail() for the last)
df.info()         # columns, non-null counts, types
df.describe()     # count, mean, min, max of number columns

Calculated columns

df["amount"] = df["qty"] * df["rate"]
df["month"] = df["date"].dt.month

No filling down: the calculation applies to every row at once.

Picking columns and rows

df["customer"]                         # one column (a Series)
df[["customer", "amount"]]             # several columns (a DataFrame)
df[df["product"] == "Laptop"]          # rows where product is Laptop
df[(df["amount"] > 100000) & (df["region"] == "North")]   # AND: &, OR: |, brackets required
df[df["region"].isin(["North", "East"])]
df.sort_values("amount", ascending=False).head(5)          # top 5 orders

Quick answers

df["amount"].sum()
(df["product"] == "Laptop").sum()      # count of True values
df["customer"].nunique()               # distinct customers
df["region"].value_counts()            # rows per region

Where beginners go wrong

Practice

Download the script (and data file, if listed) below into one folder. Open a terminal in that folder and run python lesson-NN.py. Compare with the expected output, then change something and run it again: that’s how Python is learned.