# thuiar/MMSA

MMSA is a unified framework for Multimodal Sentiment Analysis.

Repository: https://github.com/thuiar/MMSA
Canonical: https://ross.abutalabs.com/products/mmsa
Language: Python
License: MIT
License Family: permissive
Topics: multimodal-sentiment-analysis, multi-task-learning, dataset, framework
Last push: 2025-01-15T15:38:10+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 1, release rhythm 8, longevity 100
- inputs: {"age_days": 2325, "days_push": 595, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1041, forks 137 (observed 2026-08-28T04:03:20.593474+00:00)

## What it is
MMSA is a unified Python framework for multimodal sentiment analysis, supporting 15 MSA models and datasets like MOSI, MOSEI, and CH-SIMS. It offers both a Python API and command-line tool for training, testing, and benchmarking models with customizable multimodal features.

## Use cases
- train and compare multimodal sentiment analysis models
- run sentiment experiments on MOSI MOSEI or CH-SIMS datasets
- benchmark MSA models in a unified framework
- tune hyperparameters for multimodal sentiment models
- extract custom multimodal features for sentiment analysis
- reproduce research results for multimodal sentiment models

## When to choose
- you need a unified interface to train and compare multiple MSA models
- you want to experiment with standard multimodal sentiment datasets
- you need both Python API and CLI access for experiments
- you want reproducible baselines for multimodal sentiment research

## When to avoid
- you need general-purpose sentiment analysis on plain text only
- you need production deployment of a sentiment model rather than research experimentation
- your data is not one of the supported multimodal formats

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, nlp, audio-processing, cli
- domain: machine-learning, deep-learning, data-science
- platform: python, cli, windows
- tags: multimodal-sentiment-analysis, sentiment-analysis, mosi, mosei, ch-sims, research-framework, pytorch, natural-language-processing, linux, macos

## Member repositories
- thuiar/MMSA (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:20.593474+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:02:59.589061+00:00, confidence not recorded.
  - readme: https://github.com/thuiar/MMSA (fetched 2026-08-28T04:03:20.593474+00:00, sha 170e9ad97365)
  - registry_pypi: https://pypi.org/pypi/mmsa/json (fetched 2026-08-29T13:04:19.761488+00:00, sha 0e953002e3eb)
- Data as of 2026-08-30T08:39:29.467469+00:00.