Unit - II ML Python Libraries

 

Python Libraries for Machine Learning 

Introduction to Machine Learning Libraries

Machine Learning (ML) is a field of Artificial Intelligence that enables computers to learn patterns from data and make predictions or decisions without being explicitly programmed.

Python is the most popular language for ML because it has many powerful libraries. The most important ones are:

1. NumPy – for numerical computing

2. Pandas – for data handling and analysis

3. Matplotlib – for data visualization

4. Scikit-learn (sklearn) – for building ML models

---

1. NumPy (Numerical Python)

What is NumPy?

NumPy is a Python library used for working with arrays and numerical data.
It is faster and more efficient than Python lists.

NumPy is mainly used for:

• Mathematical operations

• Linear algebra

• Handling large datasets

Importing NumPy

import numpy as np

---

Creating Array: array()

An array is a collection of elements of the same type.

import numpy as np
arr = np.array([10, 20, 30, 40])
print(arr)

Output:

[10 20 30 40]

2D Array:

mat = np.array([[1,2,3],[4,5,6]])
print(mat)

---

Accessing Array (Indexing)

Index starts from 0.

arr = np.array([10, 20, 30])
print(arr[0])   # 10
print(arr[2])   # 30

For 2D array:

mat = np.array([[1,2,3],[4,5,6]])
print(mat[0][1])   # 2

---

Stacking & Splitting

stack()

Used to join arrays.

a = np.array([1,2])
b = np.array([3,4])
c = np.stack((a,b))
print(c)

array_split()

Used to split arrays.

arr = np.array([1,2,3,4,5,6])
newarr = np.array_split(arr, 3)
print(newarr)

---

Math Functions

a = np.array([10,20,30])
b = np.array([1,2,3])

print(np.add(a,b))        # addition
print(np.subtract(a,b))   # subtraction
print(np.multiply(a,b))   # multiplication
print(np.divide(a,b))     # division
print(np.power(a,2))      # power
print(np.mod(a,b))        # modulus

---

Statistics Functions

arr = np.array([10,20,30,40])

print(np.amin(arr))      # minimum
print(np.amax(arr))      # maximum
print(np.mean(arr))      # mean
print(np.median(arr))    # median
print(np.std(arr))       # standard deviation
print(np.var(arr))       # variance
print(np.average(arr))   # average
print(np.ptp(arr))       # range (max-min)

---

2. Pandas

What is Pandas?

Pandas is used for data manipulation and analysis.
It works with structured data like tables (rows and columns).

Two main structures:

• Series (1D)

• DataFrame (2D)

Import Pandas:

import pandas as pd

---

Series: Series()

A Series is a one-dimensional labeled array.

s = pd.Series([10,20,30])
print(s)

---

DataFrame: DataFrame()

A DataFrame is like an Excel table.

data = {
    "Name": ["Amit","Ravi","Neha"],
    "Marks": [85, 90, 88]
}
df = pd.DataFrame(data)
print(df)

---

Read CSV File: read_csv()

CSV = Comma Separated Values.

df = pd.read_csv("student.csv")
print(df)

---

Cleaning Empty Cells: dropna()

Removes rows with missing values.

df = df.dropna()

---

Cleaning Wrong Data: drop()

Used to remove specific rows or columns.

df = df.drop(2)   # removes row with index 2

---

Removing Duplicates: duplicated()

print(df.duplicated())
df = df.drop_duplicates()

---

Pandas Plotting: plot()

df["Marks"].plot()

---

3. Matplotlib

What is Matplotlib?

Matplotlib is used for data visualization (graphs and charts).

Import:

import matplotlib.pyplot as plt

---

Line Plot: plot()

x = [1,2,3,4]
y = [10,20,30,40]
plt.plot(x,y)
plt.show()

---

Labels: xlabel(), ylabel()

plt.xlabel("X Axis")
plt.ylabel("Y Axis")

---

Grid: grid()

plt.grid()

---

Bar Chart: bar()

plt.bar(x,y)
plt.show()

---

Histogram: hist()

marks = [50,60,70,80,90,85,75]
plt.hist(marks)
plt.show()

---

Subplot: subplot()

plt.subplot(1,2,1)
plt.plot(x,y)

plt.subplot(1,2,2)
plt.bar(x,y)
plt.show()

---

Pie Chart: pie()

data = [40,30,20,10]
plt.pie(data)
plt.show()

---

Save Plot as PDF: savefig()

plt.plot(x,y)
plt.savefig("graph.pdf")

---

4. Scikit-learn (sklearn)

What is Scikit-learn?

Scikit-learn is the most popular library for Machine Learning algorithms.

It provides:

• Classification

• Regression

• Clustering

• Model evaluation

---

Key Concepts in Sklearn

1. Dataset – Collection of data

2. Features – Input variables

3. Target – Output variable

4. Model – Algorithm used for learning

5. Training – Teaching model with data

6. Testing – Checking model performance

7. Prediction – Output of the model

---

Steps to Build a Model in Sklearn

Step 1: Load Dataset

import pandas as pd
data = pd.read_csv("data.csv")

---

Step 2: Separate Features and Target

X = data[["age","salary"]]   # input
y = data["purchased"]       # output

---

Step 3: Split Data

train_test_split()

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2
)

80% → training
20% → testing

---

Step 4: Choose Model

Example: Linear Regression

from sklearn.linear_model import LinearRegression
model = LinearRegression()

---

Step 5: Train Model

model.fit(X_train, y_train)

---

Step 6: Make Prediction

pred = model.predict(X_test)
print(pred)

---

Summary Table

Library	Purpose
NumPy	Numerical operations
Pandas	Data handling
Matplotlib	Data visualization
Sklearn	Machine learning models