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

Frame Semantic Transformer

Frame Semantic Transformer is a Python library that uses pretrained Hugging Face T5 models trained on FrameNet for frame-based semantic parsing. Its inference methods identify triggers, classify frames and extract frame elements from sentence text.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Frame Semantic Transformer — 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 an NLP workflow needs FrameNet-oriented parses from sentence-level text, or when comparing the supplied base and small pretrained models for frame parsing.

Who it is for

  • NLP developers adding FrameNet-based semantic parsing to a Python workflow.
  • Researchers and educators exploring transformer-based frame parsing.
  • Developers building a semantic-parsing demonstration.

Documented uses

  • Identify semantic-frame triggers and frame types in a sentence.
  • Extract the frame elements associated with detected frames.
  • Process multiple sentence strings in batches.
Before you choose

Requirements & limits

The project is published as an open-source repository under the MIT license and provides a pip installation command. Its README also links to a live demo.

  • The base model is the default; the small model is described as faster and one quarter the size, with lower F1 than base across the reported tasks.
  • The README reports evaluation for trigger identification, frame classification and argument extraction using the same documents and methodology as Open-Sesame.
  • GPU use is attempted by default when available; the constructor can explicitly select CPU or GPU use.
  • The model is loaded lazily on the first inference call unless setup() is invoked first.

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 library and parse a sentence

Set up the Python package and obtain a FrameNet-style parse for one sentence.

Before you start

  • A Python environment capable of installing the package with pip.

Steps

  1. Install the package with the command below.

    Documented command
    pip install frame-semantic-transformer

  2. Import FrameSemanticTransformer from frame_semantic_transformer and create an instance.

  3. Call detect_frames() with a single sentence.

  4. Read the returned sentence, trigger locations and frames; each frame includes its name and frame elements.

Expected outcome

A structured result containing detected semantic frames and their elements for the input sentence.

Guide limitations

  • The project recommends one sentence per string; multiple sentences in one string may not be handled reliably.
  • The model loads lazily when inference is first called unless setup() is called earlier.
Documentation 1 ↗
Source-based practical guide

Parse a batch of sentences

Process a list of sentence strings through the library’s bulk inference method.

Before you start

  • The installed Python package and its FrameSemanticTransformer class.

Steps

  1. Create a FrameSemanticTransformer instance, optionally setting batch_size.

  2. Pass a list of sentences to detect_frames_bulk().

  3. Use the returned results for the corresponding sentence parses.

Expected outcome

Frame-parsing results for a list of sentences processed in batches.

Guide limitations

  • Provide single sentences as list items; the project says paragraph text should be split into sentences first.
  • The default batch size is 8.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
chanind/frame-semantic-transformer · Official homepage title · retained in its source language
How we review sourcesSuggest a correction