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 available in two ways: as a permanently deployed web app on Streamlit Cloud that you can open in any browser right now, or as a locally-hosted app that you run from your own machine. Both options give you the identical full-featured interface — choose whichever fits your workflow. No account, sign-up, or API key is needed for either option.

Option 1 — Use the Live App

The fastest way to try TryMLEasy is to visit the hosted deployment directly. Everything runs on Streamlit’s infrastructure — nothing to install. Open your browser and navigate to:
https://trymleasy-2-0.streamlit.app/
The app loads immediately. You can upload any numerical CSV file and work through the full ML pipeline right away. This is the recommended path if you just want to explore the tool before committing to a local setup.

Option 2 — Run Locally

Running TryMLEasy locally gives you full control over the source code and lets you experiment with or extend the app. You’ll need Python 3.8 or newer and pip installed on your system.
1

Clone the repository

git clone https://github.com/Gurneet1928/TryMLEasy.git
2

Enter the project directory

cd TryMLEasy
3

Install all dependencies

pip install -r requirements.txt
This installs Streamlit, scikit-learn, TensorFlow, Keras, Plotly, Pandas, seaborn, Matplotlib, and all supporting libraries. TensorFlow can take a few minutes to download.
4

Launch the app

streamlit run home_page.py
5

Open the app in your browser

Streamlit will print a local URL in your terminal. Open it in any browser:
http://localhost:8501
The TryMLEasy home page will appear with a CSV file uploader ready to go.

Train Your First Model

Once the app is open — whether via the cloud URL or your local server — follow these steps to go from raw CSV to trained model results.
The repository includes a ready-made sample_data.csv file in the repo root. Use it to follow this walkthrough immediately without needing your own dataset.
Only CSV files with numerical columns are currently supported. If your dataset contains text or categorical columns, encode them numerically before uploading.
1

Upload your CSV dataset

On the Main Page, click the file uploader and select your .csv file (or use sample_data.csv from the repo root for a quick test). Once a file is selected, the Next button becomes active. Click it to proceed.
2

Review your dataset

The Show Dataset page displays your full dataframe so you can verify it loaded correctly — column names, row counts, and values are all visible. Click Next (Feature Selection) when you’re satisfied.
3

Configure feature selection

On the Feature Selection page:
  • Select the feature columns you want to use as model inputs (multi-select).
  • Choose the target column — the variable the model will predict.
  • Optionally pick a scaler (Standard Scaler, MinMax Scaler, Robust Scaler, or Normalization) to preprocess your features.
  • Use the test ratio slider to set the proportion of data held out for evaluation (e.g., 0.2 for an 80/20 train-test split).
Click Next once configured.
4

Choose Traditional ML Model

On the Model Selection page, select Traditional ML Model as your model type and click Next.
5

Select and train a model

On the Traditional ML Model page:
  • Choose Classification or Regression depending on your target variable.
  • Pick a specific model from the dropdown (e.g., Logistic Regression for classification, or Linear Regression for regression).
  • Click Train Model.
TryMLEasy will split the data, fit the model, and run predictions on the held-out test set.
6

Review your results

After training, the results page displays:
  • Evaluation metrics — Accuracy, F1 Score, Precision, and Recall for classification; MSE, Explained Variance, and Max Error for regression.
  • Predicted vs. True scatter plot — an interactive Plotly chart comparing the model’s predictions against actual target values on the test set.

Build docs developers (and LLMs) love