Skip to content

Latest commit

History

26 Commits

Folders and files

NameName
Last commit message
Last commit date

Repository files navigation

🔬 Multi-Algorithm Classifier Comparison App

Interactive Streamlit app benchmarking SVM, Logistic Regression, and Perceptron across 3 real-world Kaggle datasets — spanning binary classification (fruit, pumpkin) and multi-class classification (9 fish species) — with Grid Search hyperparameter tuning and real-time algorithm/dataset selection.

🔗 Live Demo:Try it here

📊 Results

DatasetTaskBest Accuracy / CV Score
FruitBinary classification96.5% test accuracy
FishMulti-class (9 species)94.0% CV score
PumpkinBinary classification89.0% CV score

🧠 Methodology

  • Extensive Grid Search hyperparameter tuning per algorithm-dataset pair (kernel, C, gamma, solver, regularization)
  • Users can select any algorithm/dataset combination and instantly view classification results

📂 Project Structure

App/ # Streamlit application + saved models
SVM/ # SVM training notebooks per dataset
Logistic Regression/ # Logistic Regression training notebooks per dataset
Perceptron/ # Perceptron training notebooks per dataset

⚙️ Tech Stack

Python · Scikit-learn · Streamlit

🚀 Running Locally

cd App
pip install -r requirements.txt
streamlit run app.py

About

Interactive Streamlit app comparing SVM, Logistic Regression, and Perceptron across 3 real-world Kaggle datasets (fruit, pumpkin, fish) with Grid Search hyperparameter tuning.

Topics

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages