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.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.
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.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.
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.
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.
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.