This post briefly discusses the newest feature introduction to this site: a comprehensive, no-code, drag&drop interface for training and comparing machine learning models on disparate datasets

🚀 No-Code Machine Learning Pipeline

Interactive · Browser-Based · Zero Install

Build, train, and evaluate machine learning models directly in your browser — no code required.

🌐 Live Demo

🔗 Try it here: Click Me!

🧠 What This Tool Does

• 📂 Upload your dataset (CSV)
• 🔍 Perform Exploratory Data Analysis (EDA)
• ⚙️ Apply feature engineering & preprocessing
• 🤖 Train multiple ML models
• 📊 Evaluate performance (ROC, AUC, F1, Confusion Matrix)
• 🔮 Generate predictions on new data

⚡ Key Features

• 💻 100% Browser-Based (no setup)
• 🧩 Step-by-step guided workflow
• 📊 Built-in visualizations
• 🧠 Multiple ML algorithms
• 📈 Real-time evaluation

🏗️ Supported Models

• Logistic Regression
• K-Nearest Neighbors
• Gaussian Naive Bayes
• Decision Tree
• Random Forest
• ADA Boost
• Gradient Boosting Machine

🪜 Step-by-Step Usage

📥 1. Load Dataset

• Upload CSV
• Ensure binary target column
• Index columns auto-removed

🎯 2. Select Target & Features

• Choose binary target
• Define positive class
• Configure feature types

🔍 3. Explore Data

• Summary stats
• Class balance
• Visualizations

⚙️ 4. Preprocess Features

• Apply scaling / transformations

🎛️ 5. Configure Models

• Select models
• Adjust split

🏋️ 6. Train

• Run pipeline
• Monitor progress

📊 7. Results

• Metrics + visualizations

🔮 8. Predict

• Upload new data
• Generate predictions

🛠️ Tech Stack

• Vanilla JavaScript
• Chart.js
• PapaParse

⭐ Support

Star the repo if you find it useful!