OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Tron

Tron is a Python implementation of a session-based transformer recommender, built with PyTorch Lightning and designed for recommendation experiments using optimized negative sampling.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Tron — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Tron when you want a code-based research starting point for session-based recommendations and need configurable experiments involving datasets, negative sampling and loss functions.

Who it is for

  • Recommender-system researchers exploring session-based transformer models.
  • Developers prototyping sequential recommendations for e-commerce catalogs.

Documented uses

  • Train a session-based recommendation model from a configured dataset.
  • Prepare a supported dataset for an experiment.
  • Compare model, sampling and loss-function settings through experiment configurations.
Before you choose

Requirements & limits

The public GitHub repository is listed under the MIT license.

  • The project is a Python code repository, not a hosted recommendation service.
  • PyTorch and PyTorch Lightning are the documented framework technologies.
  • Configuration files expose choices such as dataset, model, negative-sampling style, loss, optimizer, training device and maximum session length.
  • The quick-start instructions include system installation of 7zip, gzip and unzip utilities.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare a dataset and launch a configured experiment

Set up the repository, prepare a named dataset and start an experiment using its configuration file.

Before you start

  • Python environment with pip available
  • 7zip, gzip and unzip utilities
  • A dataset supported by the repository, such as Yoochoose

Steps

  1. Clone the repository from its GitHub URL.

  2. Install Pipenv, then install the project’s development dependencies with pipenv install --dev.

  3. Install 7zip, gzip and unzip utilities on the system.

  4. Prepare Yoochoose with ./prepare.sh yoochoose.

  5. Start the configured experiment with pipenv run python -m src --config-filename tron/yoochoose.

Expected outcome

The repository’s main script is invoked with the `tron/yoochoose` configuration.

Guide limitations

  • The README provides this example for Yoochoose; it does not document dataset-specific preparation details for every listed dataset.
  • The README lists CPU and GPU as configuration device options but does not establish hardware requirements or performance.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
otto-de/TRON · Official homepage title · retained in its source language
How we review sourcesSuggest a correction