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

Lednet

Lednet is a public Python repository containing PyTorch code for LEDNet, a lightweight encoder-decoder neural network for real-time semantic segmentation. Its ResNet-based encoder uses channel split and shuffle operations, while its decoder uses an attention pyramid network. The repository includes training and testing code, with Cityscapes as its documented supported training dataset.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Lednet — 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 Lednet when you need the repository’s PyTorch implementation and documented Cityscapes training or evaluation workflow, or want to study its lightweight encoder-decoder design.

Who it is for

  • Computer-vision developers evaluating a lightweight semantic-segmentation model.
  • Researchers and learners studying LEDNet or reproducing a Cityscapes-oriented training and evaluation workflow.

Documented uses

  • Train LEDNet on the documented Cityscapes dataset.
  • Evaluate segmentation results with the included IoU tools.
  • Generate colorized test-result images and measure forward time using repository scripts.
Before you choose

Requirements & limits

The repository is public and specifies the MIT license.

  • The README describes a legacy environment: Python 3.6, PyTorch 0.4.1, CUDA 9.0 and cuDNN 7.1.
  • Training requires Cityscapes images and annotations, with a documented directory layout and label-conversion step.
  • The repository’s performance figures refer to stated experimental conditions and should not be assumed for different hardware.

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 the documented Cityscapes training setup

Install the listed dependencies and arrange Cityscapes data for the repository’s training workflow.

Before you start

  • Python 3.6.x; the README describes PyTorch 0.4.1, CUDA 9.0 and cuDNN 7.1.
  • Cityscapes images and annotations, plus CityscapesScripts for label conversion.
  • Visdom and torchsummary.

Steps

  1. Clone the repository and enter its project directory.

  2. Install the Python dependencies with the command below.

    Documented command
    pip3 install -r requirements.txt

  3. Install Visdom and torchsummary.

  4. Download the Cityscapes image and annotation packages and arrange the data in the documented directory structure.

Expected outcome

The repository and Cityscapes data are prepared for the documented training stage.

Guide limitations

  • This preparation guide does not establish that training or inference has been run.
  • The README names Cityscapes as the currently supported training dataset; CamVid, VOC and ADE20K are described as intended additions.
Documentation 1 ↗
Evidence and transparency

Sources & review

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