Importing Data — Class Starter

This repository contains a demo and practice set for a short lesson on importing different data types into Python (primarily with Pandas).

Structure

importing-data-class/
├─ data/
│  ├─ demo/            # instructor-led
│  └─ practice/        # student exercise
├─ notebooks/
│  ├─ 01_demo.ipynb
│  └─ 02_exercises.ipynb
├─ src/
│  └─ util.py
├─ requirements.txt
└─ README.md

Setup

uv init
uv add ipykernel pandas lxml openpyxl

Demo outline

1. CSV/TSV — pd.read_csv with sep, usecols, dtype, nrows, compression.
2. Excel — pd.read_excel one sheet vs sheet_name=None for all sheets.
3. JSON — Python's json + pd.json_normalize for nested data.
4. Plain text — open() / context manager & .read() vs .readlines().
5. HTML tables — pd.read_html() returns a list of DataFrames; select one by index or with match= / attrs=.

Practice tasks (10–15 min)

Open notebooks/02_exercises.ipynb and complete the TODO cells:

• Load practice/tsv/air_quality.tsv as a DataFrame (tab-delimited), set date as the index.
• Read practice/json/events.json, flatten attendees to one row per person.
• Count lines that contain "ERROR" in practice/text/log.txt.
• Read all tables from practice/html/wiki_table.html and select the one with id awards.
• Load practice/csv_gz/movies.csv.gz (gzipped) and compute the average imdb_rating.
• (Optional) Read practice/excel/sales_regions.xlsx and join both sheets on region.

Notes

• If Excel engines are unavailable in your environment, the Excel files may be omitted; install openpyxl to enable reading/writing .xlsx.
• Everything is small and self-contained so it works without internet access.