OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Recnn

RecNN is a Python toolkit built around PyTorch for developing reinforcement-learning recommendation systems, with a focus on personalized news recommendation and online off-policy learning using dynamically generated item embeddings. Developers can use complete algorithms or work with network and learning components separately to assemble custom recommendation workflows.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Recnn — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use RecNN when you want a Python/PyTorch toolkit for experimenting with reinforcement-learning recommenders and need control over algorithms or their constituent components.

Who it is for

  • Developers prototyping personalized news or other recommendation models with reinforcement learning.
  • Researchers experimenting with off-policy recommendation algorithms and model components.
  • Python users working with sequential or fixed-frame recommendation data.

Documented uses

  • Build a reinforcement-learning-based personalized recommendation prototype.
  • Experiment with online off-policy learning and dynamically generated item embeddings.
  • Choose between using a complete algorithm and composing network and learning components.
  • Explore how user ratings affect movie recommendations through the Streamlit example.
Before you choose

Requirements & limits

The public repository lists an Apache-2.0 license and documents installation from Git with pip.

  • The toolkit is built around PyTorch 1.7, as stated in the repository description.
  • The project supports different levels of abstraction, from importing a full algorithm to defining components separately.
  • The README describes parallel data loading with Modin and TensorBoard visualization.
  • Some listed capabilities, including dynamic-size sequential data and LSTM/RNN/GRU sequential state representation, are marked work in progress; BCQ is listed as not implemented yet.
  • The Streamlit movie demo is distinct from the underlying developer library and illustrates rating-responsive recommendations.

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

Launch the RecNN Streamlit demonstration

Run the repository’s movie-recommendation demo.

Before you start

  • Git and Streamlit available in the environment.
  • A shell in which to clone and enter the repository.

Steps

  1. Clone the repository with the command below.

    Documented command
    git clone git@github.com:awarebayes/RecNN.git

  2. Change into the cloned directory with cd RecNN.

  3. Run streamlit run examples/streamlit_demo.py.

  4. Use the demo’s movie-rating feature to see how recommendations and scores respond to ratings.

Expected outcome

The README describes an interactive movie recommendation demo with movie ratings, recommendation scores, and options to inspect diversity, correlations, pairwise distances, and pinpoint accuracy.

Guide limitations

  • The README does not specify how to install Streamlit or other dependencies for the demo.
  • The described score behavior is an example from the README, not a verified result.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
awarebayes/RecNN · Official homepage title · retained in its source language
How we review sourcesSuggest a correction