Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/Gurneet1928/TryMLEasy/llms.txt

Use this file to discover all available pages before exploring further.

TryMLEasy is an open-source Python web application that removes the coding barrier from machine learning. Upload a CSV dataset, configure preprocessing, pick a model, and get evaluation results — all through an intuitive point-and-click interface powered by Streamlit. Whether you want to try classical algorithms like Logistic Regression or build a custom neural network with multiple Dense layers, TryMLEasy handles the pipeline for you.

Quickstart

Run TryMLEasy in minutes — use the live demo or spin it up locally.

Installation

Clone the repo and run the app locally with Python 3.8+.

Upload a Dataset

Learn how to prepare and upload your CSV dataset.

Traditional Models

Train classification and regression models with one click.

Neural Networks

Design and train custom neural networks with a visual layer builder.

Reference

Full reference for preprocessing, models, and evaluation metrics.

How It Works

TryMLEasy guides you through a structured, step-by-step workflow — no coding needed at any stage.
1

Upload your dataset

Provide a CSV file on the main page. TryMLEasy reads it into memory and displays a preview so you can verify it loaded correctly.
2

Select features and preprocessing

Choose your feature columns and target column. Optionally apply a scaling technique (Standard Scaler, MinMax, Robust, or Normalization) and a dimensionality reduction step (PCA, Kernel PCA, or FastICA). Set your train/test split ratio using the slider.
3

Choose and train a model

Select Traditional ML Model or Neural Network. For traditional models, pick classification or regression and choose from 10 available algorithms. For neural networks, define layers interactively and set epochs and batch size.
4

Evaluate results

View evaluation metrics (accuracy, F1, MSE, etc.) and a predicted-vs-true scatter plot. For neural networks, inspect the training vs. validation loss curve across all epochs.

What’s Included

10 ML Models

6 classification and 4 regression algorithms from scikit-learn.

4 Scalers

Standard, MinMax, Robust, and Normalization preprocessing options.

3 Decomposition Methods

PCA, Kernel PCA, and FastICA for dimensionality reduction.

Neural Network Builder

Dynamic layer table with neuron count and activation function per layer.

7 Evaluation Metrics

Full suite of metrics for both classification and regression tasks.

Free Cloud Deploy

Runs on Streamlit Cloud — accessible to anyone with a browser.
TryMLEasy currently supports numerical features only. Categorical feature encoding is planned for a future release. Make sure your CSV contains numeric columns before uploading.

Build docs developers (and LLMs) love