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

Perceiver

Perceiver is a Python package implementing the Perceiver machine-learning model in TensorFlow. It uses cross-attention to ingest data into a set of latent units and iterative attention over those units, with an architecture intended to process large and varied inputs without all-to-all Transformer attention.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Perceiver — 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 to explore or adapt a TensorFlow implementation of Perceiver when your work involves large or varied input sets and you want to examine a latent attention bottleneck.

Who it is for

  • Machine-learning practitioners exploring attention architectures in TensorFlow.
  • Researchers comparing latent iterative attention with Transformer approaches.
  • Developers adapting a Perceiver implementation for image-shaped inputs.

Documented uses

  • Install the package in a Python environment.
  • Configure a Perceiver model with input, latent, attention and class-count parameters.
  • Explore the documented image example, which passes an image-shaped tensor to a 1,000-class model.
Before you choose

Requirements & limits

  • The implementation is a Python package built on TensorFlow.
  • The example exposes model settings including input axes and channels, latent count and dimension, attention heads, and number of classes.
  • The README provides installation and development setup instructions; it does not document a custom-dataset training workflow.

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 a development checkout

Prepare a local checkout with the package's development dependencies.

Before you start

  • A virtual environment.
  • Git and pip.

Steps

  1. Clone https://github.com/Rishit-dagli/Perceiver.git.

  2. Change into the perceiver directory.

  3. Run the command below.

    Documented command
    pip install -e .[dev]

Expected outcome

The package is installed in editable mode with the development tools specified by the project.

Guide limitations

  • This setup guide does not document how to run the development or test tools.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
18 October 2026
Original website title
GitHub - Rishit-dagli/Perceiver: Implementation of Perceiver, General Perception with Iterative Attention · GitHub · Official homepage title · retained in its source language
How we review sourcesSuggest a correction