OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Fired Up

Fired Up is a Python package implementing deep reinforcement-learning algorithms in PyTorch, with command-line tools to train, test and plot policies in OpenAI Gym environments.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Fired Up — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Fired Up when you want a local Python command-line workflow for experimenting with deep-RL algorithms in fully observed, non-image-based Gym environments.

Who it is for

  • Learners exploring deep reinforcement learning.
  • Developers training and evaluating policies in Gym environments.

Documented uses

  • Train policies with VPG, TRPO, PPO, DQN, DDPG, TD3 or SAC.
  • Test a trained policy and view a video of its behavior.
  • Plot results from a policy-training run.
Before you choose

Requirements & limits

The repository is public and lists an MIT license. The README identifies Linux and macOS as supported platforms; Windows is untested.

  • Requires Python 3, PyTorch, OpenAI Gym and OpenMPI.
  • MuJoCo environments are excluded from the default Gym installation.
  • The documented actor-critics are non-recurrent and intended for fully observed, non-image-based environments.

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 Fired Up and check a PPO run

Set up the package and run the README's PPO installation check in LunarLander-v2.

Before you start

  • Python 3, PyTorch, OpenAI Gym and OpenMPI.
  • Linux or macOS is identified as supported; the README describes Python 3.7 with Anaconda or a Python 3 virtual environment.
  • OpenMPI development files on Ubuntu, or Homebrew to install OpenMPI on macOS.

Steps

  1. Create and activate a Python environment, using the documented Anaconda or virtualenv approach.

  2. Install OpenMPI: the README gives sudo apt update && sudo apt install libopenmpi-dev for Ubuntu and brew install openmpi for macOS.

  3. Clone https://github.com/kashif/firedup.git, change into the firedup directory, and run the command below.

    Documented command
    pip install -e .

  4. Run the command below.

    Documented command
    python -m fireup.run ppo --hid "[32,32]" --env LunarLander-v2 --exp_name installtest --gamma 0.999

Expected outcome

The README presents a PPO training run in LunarLander-v2 as the installation check.

Guide limitations

  • The README identifies Linux and macOS as supported; Windows is untested.
  • Gym MuJoCo environments are not included in the default Gym installation.
Documentation 1 ↗
Source-based practical guide

Test and plot a saved policy

Use the documented utilities on the policy output from the `installtest` run.

Before you start

  • A completed PPO run using the installtest experiment name.

Steps

  1. Run the command below to watch a video of the trained policy.

    Documented command
    python -m fireup.run test_policy data/installtest/installtest_s0

  2. Run the command below to plot the results.

    Documented command
    python -m fireup.run plot data/installtest/installtest_s0

Expected outcome

The documented utilities provide a video view of the trained policy and a plot of its results.

Guide limitations

  • These commands depend on the output path produced by the README's installtest example.
Documentation 1 ↗
Evidence and transparency

Sources & review

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