🟣 ML + AI · Lesson 60

Cross-Validation and Hyperparameter Tuning

Cross-Validation और Hyperparameter Tuning

What is Cross Validation and Hyperparameter Tuning?

Cross Validation and Hyperparameter Tuning means cross validation tests a model on multiple splits, and hyperparameter tuning finds better model settings.

In real programs, this topic helps in testing model on multiple splits. Learn the idea first, then type the program yourself and compare the output.

💡 At a Glance

Point	Details
Course Area	Machine Learning + AI Concepts used for prediction, classification, clustering and AI-based projects.
Main Use	testing model on multiple splits
Example File	`cross-validation-tuning.py`
Practice Focus	Run, change values, and explain the output line by line.

Why should you learn this?

It is useful for testing model on multiple splits.
It connects with finding better settings.
It improves your ability to read, write and debug Python programs.

Important Terms

These terms are used directly in this lesson. Understand them before memorising the code.

Term	Meaning
k-fold	Cross-validation method that tests model on multiple splits.
GridSearchCV	Tool that tries many hyperparameter combinations.
hyperparameter	Setting chosen before training, such as tree depth or k value.
validation	Checking whether data follows the required format.
overfitting	Model learns training data too closely and performs poorly on new data.

Syntax / Basic Pattern

The simple pattern is: prepare data, apply the concept, then show the result.

Basic Pattern

from sklearn.datasets import load_iris
from sklearn.model_selection import cross_val_score
from sklearn.tree import DecisionTreeClassifier
X, y = load_iris(return_X_y=True)
model = DecisionTreeClassifier(random_state=0)
scores = cross_val_score(model, X, y, cv=5)
print("Scores:", scores)
print("Average:", scores.mean())

Complete Example Program

Python – cross-validation-tuning.py

from sklearn.datasets import load_iris
from sklearn.model_selection import cross_val_score
from sklearn.tree import DecisionTreeClassifier

X, y = load_iris(return_X_y=True)
model = DecisionTreeClassifier(random_state=0)
scores = cross_val_score(model, X, y, cv=5)
print("Scores:", scores)
print("Average:", scores.mean())

Expected Output

Cross-validation scores and average accuracy will be displayed.

Program Explanation

from sklearn.datasets import load_iris imports ready-made features from a module/library.
from sklearn.model_selection import cross_val_score imports ready-made features from a module/library.
from sklearn.tree import DecisionTreeClassifier imports ready-made features from a module/library.
X, y = load_iris(return_X_y=True) stores a value in X, y.
model = DecisionTreeClassifier(random_state=0) stores a value in model.
scores = cross_val_score(model, X, y, cv=5) stores a value in scores.
print("Scores:", scores) displays information or calculated result on the screen.

Where will you use it?

Testing model on multiple splits.
Finding better settings.
Reducing random result dependence.

Common Mistakes

Training and testing the model on the same data.
Using an algorithm without understanding the input features.
Reporting only accuracy without checking actual mistakes and limitations.

Practice Tasks

Type the program in cross-validation-tuning.py and run it.
Change input values or sample data and observe the new output.
Create one example related to testing model on multiple splits.
Write 5 lines explaining the logic in your own words.

Summary

Cross Validation and Hyperparameter Tuning is not a theory-only topic. You should be able to explain the meaning, write the example, run it successfully, and use it in a small practical program.

Cross Validation और Tuning क्या है?

Cross Validation और Tuning ka matlab hai: Cross validation tests a model on multiple splits, and hyperparameter tuning finds better model settings. Simple words me, ye topic practical Python programs likhne me direct use hota hai.

Is topic ko sirf definition ke liye nahi, balki testing model on multiple splits jaise real examples ke liye practice karein.

यह क्यों सीखना जरूरी है?

Ye testing model on multiple splits me kaam aata hai.
Ye finding better settings se bhi connected hai.
Isse aap code ka output aur errors better samajh paate hain.

Important Terms

Term	Meaning
k-fold	Cross-validation method that tests model on multiple splits.
GridSearchCV	Tool that tries many hyperparameter combinations.
hyperparameter	Setting chosen before training, such as tree depth or k value.
validation	Checking whether data follows the required format.
overfitting	Model learns training data too closely and performs poorly on new data.

Syntax / Basic Pattern

Basic idea: pehle data तैयार करें, phir Python logic apply करें, aur finally result display करें.

Basic Pattern

from sklearn.datasets import load_iris
from sklearn.model_selection import cross_val_score
from sklearn.tree import DecisionTreeClassifier
X, y = load_iris(return_X_y=True)
model = DecisionTreeClassifier(random_state=0)
scores = cross_val_score(model, X, y, cv=5)
print("Scores:", scores)
print("Average:", scores.mean())

Complete Example Program

Python – cross-validation-tuning.py

from sklearn.datasets import load_iris
from sklearn.model_selection import cross_val_score
from sklearn.tree import DecisionTreeClassifier

X, y = load_iris(return_X_y=True)
model = DecisionTreeClassifier(random_state=0)
scores = cross_val_score(model, X, y, cv=5)
print("Scores:", scores)
print("Average:", scores.mean())

Expected Output

Cross-validation scores and average accuracy will be displayed.

Program Explanation

from sklearn.datasets import load_iris imports ready-made features from a module/library.
from sklearn.model_selection import cross_val_score imports ready-made features from a module/library.
from sklearn.tree import DecisionTreeClassifier imports ready-made features from a module/library.
X, y = load_iris(return_X_y=True) stores a value in X, y.
model = DecisionTreeClassifier(random_state=0) stores a value in model.
scores = cross_val_score(model, X, y, cv=5) stores a value in scores.
print("Scores:", scores) displays information or calculated result on the screen.

Practical Uses

Testing model on multiple splits.
Finding better settings.
Reducing random result dependence.

Common Mistakes

Training and testing the model on the same data.
Using an algorithm without understanding the input features.
Reporting only accuracy without checking actual mistakes and limitations.

Practice Tasks

Program ko cross-validation-tuning.py file me type karke run karein.
Values change karke output compare karein.
testing model on multiple splits par ek छोटा example banayen.
Logic ko apne words me 5 lines me likhein.

सारांश

Cross Validation and Hyperparameter Tuning ko tab complete maanenge jab aap iska meaning, example, output aur practical use clearly explain kar saken.

← Back to Python Tutorial