head()Look at the first rows.
Session 2
Exploring data with Pandas
Loading • Inspecting • Filtering
• Asking questions
Bloomberg × Asian College of Journalism
Quick check-in
Before we begin
- Did everyone manage to install Python?
- VS Code?
- Jupyter Notebook?
Common installation issuepip missing
No module named pip
Python
↓
pip
↓
Jupyter
Sometimes Python installs successfully but pip isn't available.
python --version
↓
python -m pip --version
↓
python -m ensurepip --upgrade
↓
python -m pip install notebook
Load data
↓
Inspect it
↓
Ask a question
↓
Check the answer
Pandas is less about memorising commands and more about recognising patterns.
A dataframe is just a table
Table
stored inside
df
df = pd.read_excel(...)
ObjectMemory
df
│
└── our dataset
Every command starts with df because that's where the dataset lives.
Inspect before analysing
tail()See the end of the table.
shapeCount rows and columns.
info()Check types and missing data.
Filtering is a checklist
Stage 1table
State TFR
Kerala 1.7
Bihar 2.9
Tamil Nadu 1.6
UP 2.7
Goa 1.5
Stage 2question
TFR < 2 ?
Stage 3keep True
True
False
True
False
True
Resultrows kept
Kerala
Tamil Nadu
Goa
Filtering is simply creating a True/False checklist and keeping the rows marked True.
Reading Pandas like English
CodeMeaning
df.head()
Take the dataframe and show me the first five rows.
df.sort_values(...)
Take the dataframe, sort it, show the result.
df[df["x"] > 5]
Take the dataframe. Keep rows where x is greater than 5.
Debugging isn't failure
Wrong column name.
Python doesn't recognise the name. Usually missing quotes or variable.
Check brackets. Quotes. Parentheses.
Your code worked. No rows matched the condition.
Remember
Load
↓
Inspect
↓
Question
↓
Verify