Do I need to install Anaconda?

No. A standard Python installation with a virtual environment is enough and is lighter. Anaconda and Miniconda are fine alternatives if you prefer them, and the code in this course works in both.

How much math do I need?

Basic algebra and comfort with averages and percentages are enough to begin. The course introduces the statistics you need as you go.

Data Science with Python Module 1: What Data Scientists Do and Setting Up Python

Introduction

Data science turns raw data into decisions. A data scientist asks a question, collects and cleans data, explores it, builds a model or analysis, and explains the result. This first module of the eight-module Data Science with Python course explains that workflow and sets up the tools you will use throughout.

Prerequisites

Basic Python (variables, lists, functions)
A computer where you can install software

Step-by-Step Guide

Step 1: Learn the data science workflow

Almost every project follows the same loop: define the question, get the data, clean it, explore it, model it, evaluate the result, and communicate it. Cleaning and exploring usually take most of the time, and the loop repeats as you learn more. Data science differs from data engineering, which builds the pipelines that deliver reliable data, and from analytics, which focuses on reporting what happened.

1. Ask a clear question        (what decision will this support?)
2. Collect the data           (files, databases, APIs)
3. Clean the data             (missing values, duplicates, wrong types)
4. Explore and visualize      (patterns, outliers, relationships)
5. Model                      (statistics or machine learning)
6. Evaluate                   (does it work on new data?)
7. Communicate and deploy     (charts, a report, or a service)

Step 2: Create a project folder and a virtual environment

A virtual environment keeps each project's packages separate. Create one inside a new folder and activate it. On Windows PowerShell the activation script is Scripts\Activate.ps1.

mkdir ds-course
cd ds-course
python -m venv .venv

# macOS or Linux
source .venv/bin/activate
# Windows PowerShell
.venv\Scripts\Activate.ps1

Step 3: Install the core libraries

NumPy provides fast arrays, pandas provides tables of data, matplotlib draws charts, SciPy adds statistics, and scikit-learn provides machine learning. JupyterLab gives you notebooks, where code, output, and notes live together.

pip install numpy pandas matplotlib scipy scikit-learn jupyterlab
pip freeze > requirements.txt

Step 4: Start Jupyter and run your first cells

Run jupyter lab, create a notebook, and run a cell with Shift and Enter. Visual Studio Code can also open notebooks if you prefer an editor. Check that the libraries import and print their versions.

# in a terminal
jupyter lab

# in the first notebook cell
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import sklearn

print(np.__version__, pd.__version__, sklearn.__version__)

Step 5: Make a first tiny dataset

You do not need a download to start. Build a small DataFrame in code and look at it. Every later module uses data that is either built in code like this or bundled with scikit-learn, so everything runs offline.

import pandas as pd

sales = pd.DataFrame({
    "product": ["tea", "coffee", "tea", "juice", "coffee"],
    "units": [10, 7, 12, 5, 9],
    "price": [2.5, 3.0, 2.5, 4.0, 3.0],
})
sales["revenue"] = sales["units"] * sales["price"]
sales

FAQs

Do I need to install Anaconda?: No. A standard Python installation with a virtual environment is enough and is lighter. Anaconda and Miniconda are fine alternatives if you prefer them, and the code in this course works in both.
How much math do I need?: Basic algebra and comfort with averages and percentages are enough to begin. The course introduces the statistics you need as you go.

Summary

Data science is a loop of asking, cleaning, exploring, modeling, and communicating. Set up a virtual environment, install NumPy, pandas, matplotlib, SciPy, scikit-learn, and JupyterLab, and run your first notebook cells.