LOCAL-FIRST MACHINE LEARNING

Machine learning without uploading your data.

Train, compare and understand machine-learning models directly on your device. No Python setup. No coding. Your dataset stays with you.

Start Free →See how it works
No account for free useRuns in your browserPowered by scikit-learn
MyMlLabDatasetModelsPreprocessingTrainingResultsExplainSymbolic AI

Experiment Results

Regression · local browser training

BEST MODELExtra Trees
R²0.963
MAE1.82
#MODELR²MAE
1Extra Trees.9631.82
2Random Forest.9551.98
3Gradient Boosting.9472.13
PRO roadmap: y = 487.21 - 2.13·x₁ - 0.31·x₂ + 0.14·x₃
WHY MYMLLAB

From CSV to a trustworthy model.

A focused workflow for researchers, engineers, students and analysts who want results without wrestling with Python environments.

↥

Upload and inspect

Load a CSV, choose the target and inspect dataset structure before training.

◇

Compare models

Train scikit-learn algorithms and rank them using regression or classification metrics.

≡

Preprocess visually

Configure scaling, missing-value handling and categorical encoding without code.

●

Local-first privacy

Your training data stays in the browser instead of being sent to a model-training server.

⚡

Fast experiments

Use your own device for computation and build baseline models in minutes.

✦

Built to grow

PRO will add AutoML, explainability, symbolic AI, exports and later deep learning.

HOW IT WORKS

Four steps. No environment setup.

MyMlLab turns repetitive tabular machine-learning setup into a guided browser workflow.

01

Load data

Drop in a CSV dataset and choose the target.

02

Choose models

Select regression or classification algorithms.

03

Train locally

scikit-learn runs through Python in your browser.

04

Compare results

Use metrics and visualizations to choose a baseline.

ROADMAP

Classical ML first. Then much more.

We expand only where new functionality creates real value.

ML

Classical ML

Regression, classification, preprocessing and comparison.

A

AutoML

Cross-validation, parameter search and automatic pipelines.

ƒ

Symbolic AI

Discover compact equations alongside predictive black-box models.

LOCAL MODE

Your dataset stays on your device.

The free Studio is designed around browser-side computation. The website is online while training data remains local to the browser session.

Read Privacy →

Try MyMlLab on your own dataset.

Start with the free browser-based machine-learning studio.

Open MyMlLab Studio →
MyMlLab
Machine Learning Studio
PRO
IN-BROWSER MACHINE LEARNING

Build ML models without writing code.

Upload your data, choose a target and compare scikit-learn models directly in your browser.

START IN ONE CLICK

Try MyMlLab with a real dataset

Choose an example. MyMlLab loads the data and selects the correct target and task automatically.

● Runs locally

Example datasets are supplied for demonstration and education.

⇧

Drop a CSV dataset here

or select a file from your computer

● Processed locally in your browser
EXPERIMENT SETUP

Dataset configuration

No data
Waiting for a target column.
OBSERVATIONS—
COLUMNS—
MISSING—
ENGINEscikit-learn
PREVIEW

No dataset loaded

Loading ML engine…
Upload a CSV or load the demo dataset to begin.
STEP 2

Complete statistical analysis

Audit data quality, distributions, outliers, dependence, multicollinearity and target relationships before modelling.

DATASET AUDIT

No analysis yet

WAITING

Load a dataset, choose the target and run the analysis.

NUMERIC—
CATEGORICAL—
DUPLICATES—
MISSING %—
CONSTANT—
MEMORY—
DATA QUALITY

Column audit

ColumnTypeMissingMissing %UniqueRole / warning
Run statistical analysis.
NUMERICAL SUMMARY

Descriptive statistics

VariableCountMissingMeanStd.VarianceSEM95% CI low95% CI highMinQ1MedianQ3MaxIQRCV %SkewKurtosisShapiro p
Run statistical analysis.
CATEGORICAL SUMMARY

Frequencies and entropy

VariableCountMissingUniqueModeMode countMode %Entropy
Run statistical analysis.
DISTRIBUTION

Histogram, KDE, ECDF and box summary

Choose a variable after analysis.
NORMALITY

Normal Q–Q plot

Q–Q data will appear after analysis.
DEPENDENCE

Correlation heatmap

Correlation matrices will appear after analysis.
OUTLIERS

IQR and robust MAD detection

VariableIQR countIQR %MAD countMAD %
Run statistical analysis.
MULTICOLLINEARITY

VIF, tolerance and condition number

Run statistical analysis.
VariableVIFToleranceAssessment
Run statistical analysis.

Highly correlated predictor pairs (|r| ≥ 0.85)

Variable AVariable BPearson r
Run statistical analysis.
TARGET ANALYSIS

Feature–target relationships

FeatureAssociation
Run statistical analysis.
AUTOMATIC REVIEW

Warnings and recommendations

  • Run statistical analysis to generate recommendations.
STEP 3

Choose models by family

Explore scikit-learn model families and select up to three algorithms.

3/ 3 selected
ACTIVE EXPERIMENTSelected models
Choose up to three algorithms.
Grouped as in scikit-learn
STEP 4

Preprocessing & validation lab

Configure and compare up to three leakage-safe pipelines with holdout or cross-validation.

PIPELINE COMPARISON

Preprocessing configurations

Pipeline A is always active. Enable B and C to compare preprocessing under identical folds.

VALIDATION

Reliable evaluation

Regression ranking metrics

PRO
Hyperparameter optimization Automated tuning remains reserved for the paid tier.
DATA→ Split first→ 1 pipeline→ 5-fold CV→ FINAL TEST
STEP 5

Run experiment

Python and scikit-learn execute locally through WebAssembly.

EXPERIMENT

MyMlLab Experiment #001

READY
Dataset—
Target—
Task—
Models3
Pipelines1
Validation5-fold CV

Load a dataset and configure the experiment.

✓
Dataset privacy by design

The MVP does not send your CSV to a training server. Computation happens in this browser tab.

RESULTS

Model & pipeline leaderboard

Cross-validation stability and performance on the untouched final test set.

BEST MODEL

No experiment yet

Train models to populate the leaderboard.

PRIMARY—
SECONDARY—
THIRD—
RANKING

Experiment leaderboard

Waiting
RankModelScore
No results yet.
BEST MODEL

Prediction view

A chart will appear after training.
COMPREHENSIVE METRICS

All applicable evaluation metrics

ModelRun an experiment
No results yet.
✦
COMING IN PRO

Explainability, AutoML & Symbolic AI

Cross-validation, parameter optimization, feature importance, downloadable models, equation discovery and publication-ready reports.

MYMLLAB PRO

PRO is coming

This MVP contains the free browser engine. Authentication, subscriptions and protected PRO features are the next layer.

✓ Cross-validation ✓ Hyperparameter optimization ✓ AutoML ✓ Explainability ✓ Symbolic AI ✓ Full export
PRIVACY

Local-first by design.

Using a web application should not automatically mean uploading your ML dataset to someone else's training server.

Current Free Studio: dataset processing and model training are designed to happen inside the user's browser session.

What happens to your dataset?

In the current MVP, a CSV selected in Studio is read by the browser and processed by the browser-based Python environment. The MVP does not provide a dataset-upload endpoint for model training.

What does local-first mean?

The product prefers computation on the user's device whenever practical. The website and browser libraries are delivered over the internet, but the goal is to keep training data local unless a future feature explicitly requires otherwise.

Future accounts and PRO features

Future versions may use external services for authentication, subscription status, billing, optional project synchronization or optional cloud computation. Cloud computation should be a clearly distinguishable mode.

Important limitation

This page describes the intended architecture of the current MVP. Before commercial launch, a formal Privacy Policy and Terms should be completed for the actual services in use.

PRICING

Start free. Upgrade when MyMlLab starts saving you serious time.

The free version is a real ML tool. PRO is for deeper validation, optimization, explainability and reproducible outputs.

FREE

MyMlLab Free

€0 / forever
  • CSV datasets
  • Up to 5,000 rows
  • Regression and classification
  • Up to 3 models per experiment
  • Basic preprocessing
  • Train/test split
  • Core metrics
  • Local browser computation
Start Free
PRO

MyMlLab PRO

€19 / month — planned
  • Everything in Free
  • Expanded model library
  • Cross-validation
  • Grid and randomized search
  • Advanced preprocessing
  • Feature selection and PCA
  • Explainability
  • Python / notebook export
  • AutoML — planned
  • Symbolic AI — planned
Try Free Studio
Current status: the public MVP is free. PRO billing is not active yet.
DOCUMENTATION

Using MyMlLab Studio

The MVP provides a guided browser workflow for tabular regression and classification using scikit-learn.

1. Load a CSV

Open Studio and upload a CSV dataset. Free currently accepts up to 5,000 rows.

2. Choose target

Select the column to predict. MyMlLab attempts to detect regression versus classification.

3. Analyze statistics

Run the audit, distributions, dependence, outlier and target diagnostics.

4. Models & preprocessing

Compare up to three models and three leakage-safe preprocessing pipelines.

5. Validate & train

Rank combinations with holdout or cross-validation, then score the winner on an untouched final test set.

6. Compare

Use the leaderboard and visualization to compare selected models.

V13.2 Evaluation Lab

The Statistics step provides column auditing, descriptive statistics, distributions, Q-Q plots, Pearson/Spearman/Kendall correlations, IQR and MAD outliers, VIF/tolerance, target relationships and automatic data-quality recommendations.

Models are organized into scikit-learn-style families. A selected algorithm is shown with a purple highlight, glow, check mark and selected-model chip. Free experiments compare up to three models and up to three independently configured preprocessing pipelines.

Pipeline A can pass clean numerical predictors through unchanged as Original Data. Custom preprocessing includes optional numeric and categorical imputation, optional one-hot or ordinal encoding, five scaling choices, Yeo-Johnson and quantile transforms, variance/F-score/mutual-information selection and PCA. Validation supports an untouched final test set plus holdout, 3-fold, 5-fold or 10-fold cross-validation.

Regression reports R-squared, adjusted R-squared, MAE, MSE, RMSE, median and maximum error, MAPE, sMAPE, explained variance, mean error and conditional MSLE/RMSLE. Classification reports accuracy, balanced accuracy, macro and weighted precision/recall/F1/Jaccard, specificity, MCC, Cohen's kappa, losses, ROC-AUC, PR-AUC, Brier score, a counts/row-percent confusion matrix and per-class metrics.

Current Free models

V13.2 includes 33 regression algorithms and 26 classification algorithms across baselines, linear and regularized models, robust and generalized linear models, discriminant analysis, Naive Bayes, trees, ensembles, nearest neighbors, cross decomposition, neural networks, kernel methods and Gaussian processes. SVM and SVR are deliberately excluded.

One-click example datasets

Studio includes a 500-row sample of the UCI Combined Cycle Power Plant dataset for regression, the scikit-learn Breast Cancer Wisconsin Diagnostic dataset for binary classification, and the scikit-learn Iris dataset for multiclass classification. Selecting an example configures its target, task and recommended scaler automatically.

Browser requirements

A modern desktop browser is recommended. The first load can take longer because the Python runtime and scientific packages must be downloaded.

MyMlLab is an experimental MVP. Validate important results independently before using them for high-stakes decisions.
ABOUT

A simpler way to experiment with machine learning.

MyMlLab is a local-first machine-learning studio focused on turning tabular datasets into reproducible experiments without forcing users to configure Python environments or upload data to a remote training service.

What MyMlLab is trying to solve

Many experiments repeat the same setup work: importing data, identifying feature types, preprocessing, splitting, configuring models, calculating metrics and comparing results. MyMlLab turns that workflow into a guided interface.

Where the product is going

The first releases focus on classical tabular ML. Planned directions include stronger validation, hyperparameter optimization, AutoML, explainability, symbolic regression/classification, reproducible export and later browser-based deep learning where hardware allows it.

The product principle

Start without code. Keep control of your data. Leave with a model you can understand and reproduce.