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

Deepmoji

DeepMoji is an open-source deep-learning model and Python codebase for analyzing sentiment, emotion and sarcasm in text. It includes a pretrained model, vocabulary, dataset-processing code and examples for producing emoji predictions, 2,304-dimensional emotional feature vectors and transfer learning on a new dataset.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Deepmoji — 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 DeepMoji when you need a code-based model to explore sentiment, emotional tone or sarcasm in text, and can work with its documented Python and Keras setup.

Who it is for

  • Developers exploring emotion-aware text classification in applications.
  • Researchers or practitioners who want to inspect or adapt a pretrained model for emotion-related text tasks.

Documented uses

  • Extract emoji predictions from text.
  • Encode text as 2,304-dimensional emotional feature vectors.
  • Explore transfer learning on a new dataset.
  • Process and analyze datasets for reproducing the paper’s results.
Before you choose

Requirements & limits

The repository is public and lists an MIT license for the code and pretrained model. The README says the online demo is no longer available.

  • The README’s installation instructions assume Python 2.7; it cautions that Python 3 may require changes.
  • The Keras implementation requires either Theano or TensorFlow as its backend.
  • The pretrained weights are downloaded separately using a repository script.
  • Benchmark datasets are included for convenience, and the README says users must follow their original licenses.

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 DeepMoji and download its pretrained weights

Set up the repository’s documented dependencies and place the pretrained weights in the model directory.

Before you start

  • Python 2.7 and pip
  • Theano 0.9 or later, or TensorFlow 1.3 or later, configured as the Keras backend

Steps

  1. From the repository root, install the remaining dependencies with the command below.

    Documented command
    pip install -e .

  2. Run the command below to download the pretrained weights into the model directory.

    Documented command
    python scripts/download_weights.py

Expected outcome

The repository dependencies are installed and the pretrained weights are placed in the model directory.

Guide limitations

  • The README warns that Python 3 may require code changes.
  • The README describes the code as tested with Python 2.7 on Ubuntu 16.04; this setup guide does not verify execution.
Documentation 1 ↗
Evidence and transparency

Sources & review

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