No-Code Machine Learning

Build and compare predictive models without writing code.

Upload a CSV, choose the variable you want to predict, and let the platform handle preprocessing, model training, cross-validation, and performance comparison.

Designed for learning, exploration, and non-sensitive datasets. More model controls and export options are coming as the platform develops.
1

Upload your dataset

Drop a CSV here

or select a file from your computer

No dataset loaded.
2

Choose what you want to predict

The target is the outcome you want the model to learn. Examples: churn, house price, fraud, purchase outcome, or customer category.

3

Train & compare models

Exploring the CSV structure happens in your browser. Clicking Train Models sends the CSV to the hosted analysis API. Do not upload confidential, regulated, medical, financial-account, or other sensitive personal data.
4

Review your results

Best Performing Tested Model

Selected using the API's comparison metric.

Results Explained

What do these numbers mean?

The platform reports different metrics depending on whether the problem is classification or regression. Higher is not always better for every metric, so context matters.

ClassificationAccuracy shows overall correctness. Precision and recall describe different types of classification mistakes. F1 balances precision and recall.
ROC-AUCFor supported binary classification tasks, ROC-AUC measures how well a model separates the two classes across decision thresholds.
RegressionR² describes explained variance, while MAE and RMSE report prediction error. Lower MAE/RMSE is generally better.