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

Capsule Networks

Capsule Networks is an open-source Python and PyTorch implementation of a capsule neural network using iterative routing-by-agreement, with a training script configured for MNIST image recognition.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Capsule Networks — 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 it when you want a code-based MNIST example of capsule networks and iterative routing, and can work with the listed Python and machine-learning dependencies.

Who it is for

  • Developers learning or prototyping capsule-network image classification.
  • Researchers exploring dynamic routing-by-agreement through an MNIST training example.

Documented uses

  • Train the provided capsule-network model on MNIST.
  • Adjust training and routing settings in the script.
  • Study or adapt the routing-by-agreement implementation for image-classification experiments.
Before you choose

Requirements & limits

The repository is publicly available and described as open source.

  • The documented training workflow targets MNIST; adapting it to other datasets is not covered by the setup instructions.
  • The README requires a running Visdom server in addition to the listed Python and machine-learning packages.
  • The reported 99.7% accuracy is a project-reported result for a particular training run, not a performance guarantee.

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

Set up and start MNIST training

Adjust the provided training settings and start the documented capsule-network training workflow.

Before you start

  • Python 3
  • PyTorch, TorchVision, TorchNet, TQDM and Visdom
  • A running Visdom server

Steps

  1. Edit the training epochs, batch size and other settings in capsule_network.py.

  2. Start the Visdom server and run the training script using the README's documented command.

  3. Allow the script to download MNIST if it is not already in the script's directory.

Expected outcome

The training script runs the capsule-network workflow configured for MNIST.

Guide limitations

  • The repository's README describes the implementation as barebones and says complete support is not guaranteed.
  • The documented dataset setup is MNIST; extending the workflow to other datasets is listed as a TODO.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
gram-ai/capsule-networks · Official homepage title · retained in its source language
How we review sourcesSuggest a correction