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

Uniformer Pytorch

Uniformer Pytorch is a Python and PyTorch implementation of UniFormer, a video-classification model that combines attention and 3D convolutions to process spatial-temporal video tensors and return class logits.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Uniformer Pytorch — 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 need a PyTorch model implementation for processing video tensors into classification scores, rather than a hosted service that ingests and analyzes clips for you.

Who it is for

  • Python and PyTorch developers adding video classification to an application
  • Researchers or engineers working with spatial-temporal video representations

Documented uses

  • Pass a video tensor through a configurable classification model and obtain class logits.
  • Use a video-classification model as one component in a broader video-analysis application.
Before you choose

Requirements & limits

The public GitHub repository specifies an MIT license and documents installation with pip.

  • Inputs must follow the documented five-dimensional batch, channel, time, height and width layout.
  • The example exposes class count, stage dimensions, depths and local/global aggregation choices.
  • The usage instructions show inference from a supplied tensor but do not document a full video-loading pipeline.

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

Install the package and run the documented model example

Set up the Python package and pass a sample video tensor to a Uniformer-S model.

Before you start

  • Python environment with PyTorch available
  • Video tensor shaped (batch, channels, time, height, width)

Steps

  1. Install the package with the command below.

    Documented command
    pip install uniformer-pytorch

  2. Import torch and Uniformer from uniformer_pytorch.

  3. Create a Uniformer-S model with the documented class count, dimensions, depths and aggregation types.

  4. Pass a tensor shaped like the example (1, 3, 8, 224, 224) to the model and collect its logits.

Expected outcome

The example returns logits with shape `(1, 1000)`.

Guide limitations

  • The sample uses a randomly generated tensor rather than a real video-loading workflow.
  • The documented call does not explain checkpoint loading or mapping logits to application-specific labels.
Documentation 1 ↗
Evidence and transparency

Sources & review

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