Skip to content

Latest commit

 

History

History
86 lines (65 loc) · 1.91 KB

README.md

File metadata and controls

86 lines (65 loc) · 1.91 KB

🤖Machine Learning Classifier⚙️

A Python implementation of handwritten digit recognition using Support Vector Machine (SVM) and Principal Component Analysis (PCA).

Note: Due to privacy policies, I am not allowed to post the dataset publicly.

Table of Contents📋

Overview📝

This project implements a machine learning pipeline for recognizing handwritten digits using the following techniques:

  • Dimensionality reduction with PCA
  • Classification using linear SVM
  • Performance evaluation on validation set

Dependencies🛠️

from sklearn.datasets import load_digits
from sklearn.decomposition import PCA
from sklearn import svm

Dataset📂

The project uses the Digits Dataset from scikit-learn, which contains:

  • Handwritten digit images (0-9)
  • Features extracted from the images
  • Target labels indicating the digit

Implementation💻

1. Data Loading

digits = load_digits()
x, y = digits.data, digits.target

2. Dimensionality Reduction

pca = PCA(n_components=8)
pca.fit(x)
x = pca.transform(x)

3. Model Training and Prediction

svc = svm.SVC(kernel='linear')
svc.fit(x_train, y_train)
y_predicted = svc.predict(x_valid)

Running the Code▶️

  1. Load the digits dataset
  2. Apply PCA transformation
  3. Split data into training and validation sets
  4. Train SVM classifier
  5. Make predictions and evaluate performance

Experimentation⚗️

The following parameters can be modified to optimize performance:

  • Number of PCA components
  • Training set size
  • SVM kernel and parameters

Learning curves can be plotted to visualize the impact of:

  • Number of PCA components vs. accuracy
  • Training set size vs. accuracy