Importing Libraries and Loading the Dataset

Step 1: Install and import the libraries

If the libraries are not installed yet, install them once from the terminal (or a notebook cell starting with !):

pip install pandas numpy matplotlib seaborn scipy scikit-learn

Now import them in your notebook:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import warnings

warnings.filterwarnings("ignore")
sns.set_style("darkgrid")

pd, np, plt, sns are the standard short names used by almost every data scientist.
warnings.filterwarnings("ignore") hides harmless warnings so the output stays readable. Use it carefully; in a real project it is better to read the warnings first.
sns.set_style("darkgrid") gives every later plot a clean grid background.

Step 2: About the dataset

We use the WineQT dataset (red wine quality). Each row is one wine and each column is a laboratory measurement.

Column	Meaning
fixed acidity, volatile acidity, citric acid	Different acids in the wine
residual sugar	Sugar left after fermentation
chlorides	Salt content
free / total sulfur dioxide	Preservative levels
density	Density of the wine
pH	Acidity level
sulphates	Wine additive
alcohol	Alcohol percentage
quality	Score given by tasters (target variable)
Id	A row identifier (not useful for analysis)

Step 3: Read the CSV file

df = pd.read_csv("WineQT.csv")
print(df.head())

pd.read_csv() reads a comma-separated file into a DataFrame. Useful options:

Option	Use
`sep=";"`	when the delimiter is not a comma
`usecols=[...]`	load only some columns
`nrows=1000`	load only the first 1000 rows (good for huge files)
`index_col="Id"`	use a column as the row index
`na_values=["?", "NA"]`	treat these strings as missing
`encoding="utf-8"`	fix text-encoding errors

df.head() shows the first five rows. Use df.head(10), df.tail() or df.sample(5) to see more.

Common mistakes

Wrong file path - use os.getcwd() to check the current folder.
Reading Excel files with read_csv - use pd.read_excel() instead.
Forgetting that the first look at data can be misleading; a random sample() is often more honest than head() because files are often sorted.