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

Glasses

Glasses is an open-source Python library for computer-vision deep learning, with shared APIs for classification and segmentation models, pretrained weights hosted on Hugging Face Hub, and examples for customization and interpretability.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Glasses — 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 Glasses when developing Python computer-vision experiments and you want a shared interface for pretrained models, model customization or documented interpretability workflows.

Who it is for

  • Computer-vision developers comparing or prototyping with pretrained neural networks.
  • Practitioners adapting image-classification models, including changing activations, class counts or encoder training status.
  • Developers exploring model features or visual interpretations with Grad-CAM and saliency-map tools.

Documented uses

  • Load a pretrained computer-vision model and its corresponding input transformation.
  • Inspect which models are available and which are marked pretrained.
  • Customize a classification model’s activation function or class count and freeze its encoder.
  • Access internal encoder features and generate image-model interpretations.
Before you choose

Requirements & limits

The repository is public and identifies the project license as MIT. The README documents installation through pip from the Git repository.

  • The README documents PyTorch imports and Python package installation with pip.
  • Models are hosted on Hugging Face Hub; the example loads a pretrained ResNet and selects its named input transformation.
  • The library describes shared APIs across models and documents classification and segmentation.
  • Customization examples cover activation, class count and freezing the encoder; interpretability examples name Grad-CAM and SaliencyMap.

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 Glasses and load a pretrained model

Install the library and follow the documented example to load a pretrained ResNet and its matching input transform.

Before you start

  • Python environment with pip
  • PyTorch, imported by the documented example

Steps

  1. Install Glasses with pip using the Git repository URL.

  2. Import AutoModel and AutoTransform from glasses.models.

  3. Load a pretrained model such as resnet18 with AutoModel.from_pretrained and set it to evaluation mode.

  4. Select the matching transform with AutoTransform.from_name and request a summary on CPU.

Expected outcome

A pretrained model object, its named input transformation and a CPU model summary, as shown in the README example.

Guide limitations

  • The README does not provide a complete environment specification or hardware requirements.
  • This guide reflects the documented code example; it is not an execution test.
Documentation 1 ↗
Evidence and transparency

Sources & review

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