SmartML

Upload Dataset

Choose CSV File

Browse...

Header

Separator

Comma

Semicolon

Tab

Use Sample Data for Quick Exploration (10% of data)

Sample Size (%):

Dataset Preview

Training Settings

Training/Test Split Ratio:

Number of CV Folds:

Select Models to Train:

Training Mode:

Train All Selected Models

Train Single Model

Select Single Model:

Random Seed:

Hyperparameter Settings

Decision Tree Complexity Parameter:

Random Forest mtry:

SVM Sigma:

SVM Cost:

GB n.trees:

GB interaction.depth:

GB shrinkage:

Training Progress

0 / 100

Model Performance Metrics

Cross-Validation Results

Test Set Performance

Feature Importance

Select Model:

Recursive Feature Elimination

Run Recursive Feature Elimination (can be time-consuming)

Actual vs Predicted

Select Model for Predictions:

Residual Analysis

Prediction Results

About This App

SmartML: Train, Evaluate, Optimize, Tune, Select

SmartML is an interactive application designed to tackle problems with continuous target variables using advanced machine learning algorithms. It features tools for model training, evaluation, and optimization, including hyperparameter tuning, feature importance analysis, and feature selection. With these capabilities, SmartML empowers users to effectively address challenges and enhance their predictive modeling outcomes.This app implements the following algorithms:

Decision Tree (DT): A tree-based model that splits data based on feature values to predict continuous outcomes.
Random Forest (RF): An ensemble method using multiple decision trees to improve accuracy and reduce overfitting.
Support Vector Machine (SVM): A kernel-based method that finds the optimal hyperplane for regression tasks.
Gradient Boosting (GB): An ensemble technique that builds trees sequentially to minimize prediction errors.

Contact

For inquiries or support, please contact:

Developed By

M. Iqbal Jeelani (SKUAST-Kashmir)