OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Security & Safety

Nightmarenet

Nightmarenet is an open-source Python and PyTorch toolkit for adversarial-robustness training and evaluation. Its cyclic workflow moves through clean-data fine-tuning, augmentation, adversarial training and model compression or robust distillation.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Nightmarenet — OpenViST project profile in AI Security & Safety.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Nightmarenet when robustness work belongs in model training and evaluation, or when a repository needs the documented GitHub Action robustness check. Its workflow is distinct from a runtime guardrail.

Who it is for

  • Machine-learning researchers and developers training text or image-classification models.
  • Teams evaluating model performance across increasing distortion strengths.
  • Developers who want to add a robustness threshold check to a GitHub pull-request workflow.

Documented uses

  • Train a model through the Wake, Dream, Nightmare and Compress phases.
  • Apply text or image distortions as part of robustness training.
  • Evaluate clean accuracy and accuracy across increasing distortion strengths.
  • Use a GitHub Action to gate pull requests against a robustness threshold.
  • Run the documented API and frontend services locally with Docker.
Before you choose

Requirements & limits

The public GitHub repository is licensed under Apache-2.0. The documented local API-and-frontend setup uses Docker Compose.

  • The repository identifies Python 3.9+ and PyTorch 2.0+ as requirements.
  • The training workflow covers both text and image classification, with different perturbation methods for each.
  • The documented Docker setup starts the API and frontend; database, Redis and worker services are described as future hosted-platform infrastructure.
  • The repository includes notebooks for a quick start, benchmark reproduction and custom distortions.

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

Start the local API and frontend with Docker

Launch Nightmarenet’s currently supported API and frontend services locally.

Before you start

  • A local copy of the Nightmarenet repository
  • Docker Compose
  • The repository’s example environment files

Steps

  1. Copy .env.example to .env and frontend/.env.example to frontend/.env, then review and update their values.

  2. Run the command below to start the API and frontend, or use the command below to name those services explicitly.

    Documented command
    docker compose up
    Documented command
    docker compose up api frontend

  3. Consult the repository’s stack-verification instructions if you need to check whether services started healthy.

Expected outcome

The local API and frontend services are started; the documented default setup does not start database, Redis or worker services.

Guide limitations

  • The repository describes database, Redis and worker services as future hosted-platform infrastructure, not requirements for the current open-source API.
  • This setup recipe documents startup, not a verified successful run.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
Adit-Jain-srm/NightmareNet · Official homepage title · retained in its source language
How we review sourcesSuggest a correction