Getting started with Fast ML
This website allows you to automatically build and train machine learning models for regression and classification problems using your dataset in tabular format, as well as to use built models for prediction on structurally identical data.
The system performs automatic preprocessing of data, including gaps filling, encoding of categorical variables and feature selection.
You can train any of the available models and choose the one that performs best.
Trained models are saved, so you can use your models whenever you want.
Feel free to reach out to us if you want us to add more models, even ones you've written yourself!
Sign up
No need to register with a username and password, all you need is to log in with your Google account.
Supported file formats
To use the system features it is necessary to prepare data in the form of Microsoft Excel files (xlsx) or CSV files (separator should be specified by user). Extensions of training and prediction datasets should match.
Training data
- First line of the file with training data should contain feature names.
- Second and subsequent lines should contain data (one object per line).
- Training data must contain a column with the target variable, you can specify its name after downloading the file.
- The remaining columns should contain feature description of objects. The vector length must be the same for all objects.
Data for prediction
- First line of the file with data for prediction should contain feature names.
- Second and subsequent lines should contain data (one object per line).
- Table structure should be identical to the training data (column names and feature vector lengths should be the same)
- Columns should contain feature description of objects. The vector length must be the same for all objects.